UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Envisioning capitalism : geography and the renewal of Marxian political-economy Castree, Noel 1999

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1999-388646.pdf [ 20.22MB ]
JSON: 831-1.0103815.json
JSON-LD: 831-1.0103815-ld.json
RDF/XML (Pretty): 831-1.0103815-rdf.xml
RDF/JSON: 831-1.0103815-rdf.json
Turtle: 831-1.0103815-turtle.txt
N-Triples: 831-1.0103815-rdf-ntriples.txt
Original Record: 831-1.0103815-source.json
Full Text

Full Text

ENVISIONING CAPITALISM: GEOGRAPHY AND THE RENEWAL OF MARXIAN POLITICAL-ECONOMY by NOEL CASTREE B.A., The University of Oxford, 1989 M.A., The University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Geography We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October1998 © Noel Castree In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of GEOGRAPHY The University of British Columbia Vancouver, Canada D a t e 04/11/98 DE-6 (2/88) ABSTRACT Vision is the art of seeing things invisible. Jonathan Swift (1711) ... only partial perspective promises objective vision. Donna Haraway (1991) Not for the first time, Marxism is considered to be in a state of 'crisis'. This thesis seeks to 'underlabour' on behalf of a particular version of Marxism, a version articulated with force, coherence and great originality for over two decades within human geography: what David Harvey (1985a: xii), in a paradigmatic formulation, has called 'historical-geographical materialism'. A research programme, rather than the work of any one individual, historical-geographical materialism has in various ways and at various levels creatively extended the classical Marxist canon in a geographical direction. Yet today it is considered increasingly passe by critics on the Left as well as the Right of human geography, reflecting the wider ennui with Marxism outside the discipline. In particular, it is seen as being too 'modern' - too foundationalist, totalising and authoritative in its cognitive and normative claims - to contribute effectively to a critical human geography for the 1990s. Against this, this thesis seeks to develop an alternative reading of the core claims of this research programme by offering a novel reinterpretation of Marx's mature political-economy. Rewriting Marx's account of what Postone (1996: 1) calls "the fundamental core of capitalism", the thesis puts this reinterpretation of the explanatory-diagnostic basis of Marx's critique to work on three major themes of historical-geographical materialism: the production of space, the production pf nature and the production of subjectivity. It does so in order to illustrate the explanatory power, thematic reach and theoretical coherence of this reinterpretation, as well as its relevance to the late capitalist world. In closing, the normative or anticipatory-utopian basis of this reinterpreted historical-geographical materialism is considered and its political implications for today thereby scrutinised. ii TABLE OF CONTENTS Abstract ii Table of Contents iii List of Figures v Acknowledgements vi Preface vii A Note on Terminology xix PART I THE NATURE OF THE PROBLEM Chapter One Envisioning Capitalism: Marxism, Geography and the Meta- 2 Theoretical Imperative Chapter Two Explanatory and Normative Presuppositions: Harvey and 38 the Crisis of 'Traditional Marxism' PART II RECLAIMING CAPITALISM Introduction 81 Chapter Three The Limits to Theory: Science, Epistemology and 89 the Dialectical Imagination Chapter Four The Ontology of Openness: Reinterpeting Capitalism's 137 Core PART III RECLAIMING SPACE, NATURE AND THE SUBJECT Introduction 183 Chapter Five The Production of Space 191 Chapter Six The Production of Nature 229 iii Chapter Seven The Subject of Capital 259 PART IV CRITIQUE, NORM AND UTOPIA Introduction 303 Chapter Eight Theory, Politics and Normativity 311 Coda 336 Bibliography 339 iv TABLE OF FIGURES Figure 1 The Dialectic of Value Forms 122 Figure 2 The Core of Capitalism 174 Figure 3 Marx's Political-Economy and the 'Core' of Historical- 180 Geographical Materialism: Modern/Traditional and Late Modern/Traditional Figure 4 Tensions in the Space-Economy of Capitalism 211 v ACKNOWLEDGEMENTS If it is today considered rather unfashionable to express a continued interest in Marxian political-economy, then I have many people to thank for sustaining and encouraging my critical inquiries. Although we have rather different views on the limits and possibilities of Marxism(s), my greatest vote of thanks must go to Derek Gregory. His influence has, quite simply, been inestimable. In ways that I hope he will recognise, his own thinking and writing has palpably shaped the arguments I develop here. Additionally, the rare freedom he has granted me as a PhD student has allowed me to carve out my own intellectual path and for that I am deeply grateful. If Derek's influence has been crucial, then so too has that of Trevor Barnes. A devotee of political-economy, Trevor's published works, his continuing support - despite his preference for Sraffa over Marx (!) - and his good counsel have all contributed in important ways to the formulation of my ideas. I thank him sincerely. I cannot mention Derek and Trevor without also naming several other individuals in the Department of Geography at UBC. Only now do I realise what a special intellectual environment was to be found in the Department during my time there between 1990 and 1995. For ideas, arguments and social sustenance I want to thank several friends: Dave Demeritt, Steve Rice, Bruce Braun, Dan Clayton, Jock Wills, Ken Reid, Matts Jakob, Elizabeth Bronson, Kate Boyer, Michael Brown, Brett Christophers, Alison Blunt and the boys at the 'vortex'. If the Department sustained me academically and socially, then the University did so financially and I want to take this opportunity to express my enormous gratitude to UBC for providing me with five years of support, three years in the form of Graduate Fellowships and two years in the form of a Killam Pre-Doctoral Fellowship. The length and generosity of this funding still strikes me as remarkable, and I feel extremely priveleged to have been the beneficiary of it. Finally, and more recently, I must thank the Department of Geography at the University of Liverpool. Although it has taken a little longer than we both anticipated, the Department has tried to make the completion of this thesis as painless for me as it could. vi PREFACE Any really "loving" ... [theoretical] practice must fall a prey to its own critique. Gayatri Chakravorty Spivak (1990: 111) In this thesis I offer a fundamental reinterpretation of a determinate body of Marxian political-economic theory in geography. I do so in order to reclaim it for a contemporary geographical critique of capitalism. Given the current unpopularity of Marxism, both within and outside geography, this may seem an unlikely, even foolhardy endeavour. It may also appear to be a parochial one in as much as the corpus I consider has rarely penetrated to the heart of debates within the wider field of Marxism. However, since I reject both charges, I want to use these prefatory remarks as an opportunity to explain and situate my reinterpretive efforts, first in relation to Marxism, then in relation to human geography. "Marxism", as Louis Althusser (1979: 237) correctly observed, "has in its history passed through a long series of crises and transformations". However, if it is to survive as a living intellectual and political force into the twenty first century then its practitioners must respond in particularly creative ways to the present 'crisis of Marxism'. For the crisis is an especially acute one. I say this for two reasons. First, perhaps more than at any time during this century Marxism has been marginalised by an ebullient Right enjoying an unprecedented ideological hegemony. Second, this defeat by the Right has been exacerbated by Marxism's current unpopularity on the Left. Regarded as foundationalist, totalising and authoritarian - in short, as too 'modern' for these supposedly 'postmodern' times - Marxism's indifference towards and/or effacement of non-class 'otherness' and 'difference' has been meticulously exposed by postmodern, poststructural and postcolonial critics. Thus pressed on two sides, contemporary Marxists find themselves in an acute dilemma. On the one hand, the relevance of Marx's critique of political economy has in vii many respects never been greater. The gulf between the Right's ideological obfuscations ('the end of history', the triumph of liberal democracy, the beneficience ofthe 'invisible hand', the sanctity of possessive individualism and so on) and the reality of spreading poverty and intensified uneven development demands an immanent critique in which an explanatory-diagnostic account of the present conjuncture is conjoined with a sober anticipatory-utopian reading of the possibilities for progressive change. On the other hand, however, if Marxists are to thereby 'reclaim reality' (to borrow Bhaskar's [1989a] felicitous phrase) then they must deploy something like the very same claims to truthfulness, comprehensiveness and certainty that the Right has so effectively used to conceal the violent realities of a capitalist world economy and to prosecute its own counter-revolutionary case. In short, Marxism (as Marx himself realised over a century ago) will simply not be able to win over the hearts and minds of ordinary people if its cognitive claims are not based on good economic and social 'science'. But the problem is that such a strategic reversion to its 'modern' architectonics threatens to reinstall all the closures and exclusions that have exposed Marxism to so much criticism from its erstwhile comrades on the Left. Symptomatic of the difficulties of fashioning a Marxism supple enough to operate simultaneously on these two fronts is the inadequacy of many of the Marxisms presently preferred as responses to the current 'crisis'. On the one side, a number of what we might (using the simplifying language of their critics) call 'modem' Marxists working in the fields of economic theory and the history and philosophy of economic thought have bravely stuck to their guns and kept the light of classical Marxism burning through the dark days of the 1980s and 90s. Among them, one might count figures such as Chris Arthur, Alex Callinicos, Guglielmo Carchedi, Michael Eldred, Ben Fine, Norman Geras, Lawrence Harris, Ian Hunt, Joseph McCarney, Ernest Mandel, Istvan Meszaros, Fred Moseley, Patrick Murray, Bertell Oilman, Geert Reuten, Derek Sayer, Tom Sekine, Ali Shamsavari, Murray Smith, Tony Smith, Michael Williams and Ellen Meiskins Wood. viii This is, of course, a heteroclite list and I do not mean to imply any absolute identity between these authors. But what they do share, for all their differences, is a strong belief in the coherence and continued relevance of Marx's corpus in something like its original or 'classical' form. On the other side, a number of what might be called 'postmodern' Marxists have sought to open up historical materialism in the direction of otherness and difference in order to accommodate the insights of postmodern, poststructural and postcolonial critics. Most notable here has been the 'post-Marxist' work inspired by Laclau and Mouffe (1985) and the 'anti-essentialist Marxism' of Resnick and Wolff (1987), both of which present historical materialism as a non-foundationalist, non-essentialist and non-totalistic critique of political-economy.1 The strengths of each camp are the weaknesses of the other, and the mutual suspicion between 'modern' and 'postmodern' Marxists has been manifested in a series of well known and fractitious debates in which advocates on each side have been vilified by their antagonists. Modern Marxists argue that critique must be grounded in a coherent and systematic theory of political-economy, while postmodern Marxists argue that such an orthodox approach does violence to the overdetermined complexities of any social formation. Conversely, where postmodern Marxists creatively open up the borders and soften the hard edges of Marxism, modern Marxists argue that this undermines the scientific rigour necessary for theory to function effectively as critique. We seem here to arrive at something of an impasse or grand Either/Or. Either critique must rest on firm epistemological and ontological foundations even though such an absolute grounding is strictly impossible. Or critique must become honest about its partiality and precariousness but thereby relinquish the foundationalism that has allowed it to function so effectively in the past. These are, I think, debilitating alternatives. Indeed, to concede that they exhaust the current available options for any revivified historical materialism arguably amounts to an admission of the intractability of the contemporary crisis of Marxism. For this reason it seems to me vital to develop a 'third way' between what Richard Bernstein (1993: 8) aptly describes as"the Scylla of'groundless critique' and ix the Charybdis of rationally grounded critique that 'rests' on illusory foundations" if Marxism is to move forward creatively into the twenty first century. If this is possible then the phrase "the crisis of Marxism" can, as Althusser (1979: 237) argued, be given "a completely different sense from collapse and death". Instead we can say: "At last the crisis of Marxism has exploded! At last it is in fUll view! At last something can be liberated by this crisis and in this crisis". In the spirit of these stirring words, this work aims to advance the project of finding a 'third way' for Marxism. It seeks to develop a 'both/and' approach in which Marx's political-economic theory can be reappropriated as a coherent and powerful critique of the basic structures of contemporary capitalism without at the same time arrogating to itself exorbitant cognitive pretensions. Or, to put it another way, I try to show that it is possible to reinterpret the core categories of Marx's critique of political economy in the direction of what Bonefeld et al. (1995) call an "open Marxism" without, at the same time, losing any of the vitality, relevance and rigour of Marx's anatomisation of capitalist society. In so doing, I hope to contribute to a Marxism robust enough to cope with the legitimate criticisms of the non-Marxist Left while still being able to offer a systematic and compelling riposte to the political-economic hubris of the Right. If I were asked to give a summary description of the double-headed historical materialism I seek, I would - with deliberate reference to Fredric Jameson (1990)2 - call it a 'late Marxism': a Marxism neither modern nor postmodern. In this endavour I am not alone. The arguments presented here have an elective affinity with those currently being developed by scholars of Marx in the fields of political theory (e.g. Makdisi et al.'s [1996] Marxism Beyond Marxism), philosophy (e.g. Bonefled et al.'s three volume Open Marxism series [1992a, 1992b, 1995]) and literary theory (e.g. Parker's [forthcoming] Re-Marx: Reconstructive Readings in Marxist Theory and Criticism). These titles and others like them are hardly of a piece, but I am happy to note the resonances our efforts share because it signifies (at least to me) that the project to x reappropriate Marxism in a way that can negotiate the polar readings offered by modern and postmodern Marxists can be accorded at least some measure of viability. Whether my own particular effort of negotiation is a success I must leave for the reader to judge. If these comments indicate the general intent of what follows it will of course become necessary to move beyond such generalities to specify more exactly the theoretical terrain to be surveyed and reinterpreted. For clearly 'Marxism' does not "denote some fixed object with an essence to be captured if one has the necessary metaphysical skills" (McCarney 1990: viii), and the portmanteau terms 'modern' and 'postmodern' Marxism, while useful as shorthands, will hardly suffice to delineate the compass of the present work. In place of such gnomic vulgarities, then, let me offer an initial specification of the particular field to be considered. Today, far from being a tightly and clearly defined theoretical terrain, 'actually existing Marxism' - as Jameson (1996) usefully calls it - is a diffuse, variegated and heterogenous body of thought and practice. There is, then, no question of discussing Marxism tout court. Rather, we need to discriminate between multiple marxisms. Any map of these marxisms will be contentious, and one does not have to search hard for quite contrasting interpretations. However, a useful first approximation is offered by Callinicos (1989a: 1) in his recent overview of the field. He distinguishes classical or Hegelian Marxism, Althusserian or structural Marxism, and analytical or rational choice Marxism, noting that historically the latter two have been ranged explicitly against their predecessor.3 Characterised by a more or less faithful adherence to the spirit and intent of Marx's mature critique of political-economy, classical Marxism is today advocated by that relatively exclusive and beleagured coterie of authors I listed above, whom critics disparage as 'modern'. By contrast, the structural Marxism inaugurated by Louis Althusser's remarkable re-reading of Capital, although it enjoyed a career of only shortlived brilliance as a formal theoretical enterprise, has had an enduring and pervasive influence which cannot be underestimated. On the one hand, in complex and often subterranean ways it still informs xi the work of leading contemporary Marxists like Jameson, Perry Anderson and Terry Eagleton and even remains the subject of formal reconstructive efforts and applications by devotees, as in Robert Paul Resch's (1992) ambitious Althusser and the Renewal of Marxist Social Theory* On the other hand, because it pushed many of Marx's concepts to the absolute limit, Althusserian Marxism inadvertently contained the seeds of its own creative dissolution. I say 'creative', because its former advocates have been responsible for fashioning the two most ambitious and comprehensive contemporary alternatives to modern Marxism to which I have, of course, already referred: the 'post-Marxism' instigated by Laclau and Mouffe's Hegemony and Socialist Strategy (1985) and Resnick and Wolffs 'anti-essentialist Marxism', whose fullest exposition is to be found in their Knowledge and Class (1987). Finally, analytical or rational choice Marxism - drawing upon the resources of analytical philosophy and the methodologies of mainstream social science rather than what it sees as the double-speak of holistic continental philosophy - has emerged as a distinct attempt to salvage the supposed wreckage of classical Marxism by breaking it down into distinct propositions which are to be scrutinised ruthlessly, systematically and rigorously. In the substantial corpus of Jon Elster, John Roemer, Erik Wright and their co-workers, this analytical Marxism today constitues a formidable and original presence in contemporary Marxian debates.5 Perspectivising, Callinicos rightly concedes that, in the face of these two powerful rivals, classical Marxism seems exhausted as a theoretical tradition and he concludes that the renewal of Marxism as an intellectual force seems increasingly dependent on the fortunes of the structural and analytical alternatives and their offshoots. As I say, this tripartite schema is only a first approximation. As it stands it is both too tidy and insufficiently exhaustive. A fuller delineation of contemporary Marxism would also have to make reference to a series of what Gregory (1994: 102), in an apt neologism, calls "middling Marxisms", which combine historical materialism in novel and often productive ways with non-Marxian theories. These, at the very least, would xii include what Anderson (1976) called 'Western Marxism' and its legatees - that incohate and syncretic continental European tradition stretching from Lukacs, Korsch and Gramsci, through Critical Theory and French existential Marxism, to the Italian Marxism of Colletti and Delia Volpe and beyond6 - the myriad Marxist feminisms which now abound, the several extant anti-racist Marxisms (like those of Robert Miles and others), and the equally eclectic Marxisms of Habermas and fellow-travellers in the domain of contemporary social theory.7 And yet, if we supplement it in this way, Callinicos's rough map at least has the virtue of marking clearly the contemporary 'heartlands' of Marxist theory as it were, that is, the strongest options available to Marxists in these difficult times. Moreover, for my purposes this map is in at least one respect exemplary: for it indicates that even within these heartlands, never mind outside them, there is a version of historical materialism that is widely regarded as so passe as to be worthy of little further attention: classical Marxism. I say 'for my purposes' because it is precisely this version of Marxism which will be the subject of my critical energies in the chapters which follow. It thus follows that I do not concur with this sombre assessment of the theoretical resources it offers us today. It remains, of course, to say with more precision what I mean by 'classical Marxism', and I shall take up this question in due course. But answering that question depends in large part upon adding a further determination to the discussion. And it is here that my double authorial identity becomes vitally important: for if I regard myself in some sense a 'Marxist' I also most certainly consider myself a 'geographer'. Largely initiated by David Harvey's groundbreaking Social Justice and the City (1973), which was written in the wake of the civil unrest, ecological concern and stagflation of the early 1970s, 'Marxist geography' had by the mid-1980s become perhaps the best established constituent part of what we now call a 'critical human geography'. Today, however, as the Left in human geography has expanded and diversified, it no longer enjoys the pre-eminence it once took almost for granted, and rightly so: for as feminist, anti-racist, queer and green geographers xiii (among others) have shown, the geographies of power and oppression are hardly exhausted by the gyrations of capitalism. I say 'it' in a strictly improper sense because to talk of 'Marxist geography' as such is almost as question begging as it is to talk of'Marxism'. So, purposely deferring until later a discussion of the different strands of Marxist geography, let me cut to the chase. For while my object of concern is classical Marxism, it is not, in light of my disciplinary predilections, simply that. More precisely, it is that distinctive strand of Marxist geography which has extended the classical canon to tease out, coherently, consistently and with great originality, its rich but largely hidden geographical insights: what Harvey (1985a: xiv), in a memorable neologism, has called 'historical-geographical materialism'; The term is often taken to be a synonym for Harvey's work alone, and indeed it is true that Harvey remains its best known practitioner. But I want to resist this rather unthinking equation and reduction. Historical-geographical materialism is not exhausted by Harvey's oeuvre. Accordingly, although it necessarily looms very large in what follows, this thesis is not a recuperation of his work specifically - still less a Harvey hagiography. Instead, I propose that historical-geographical materialism be regarded as a research programme - involving several individuals rather than simply one - within which Harvey is the founding and most important presence. His Limits to Capital (1982), in many ways the charter document of the movement, his subsequent books (including his latest, Justice, Nature and the Geography of Difference [1996]) and his dozens of essays, constitute a formidable corpus. But in addition one must also consider the contribution of Neil Smith, whose Uneven Development (1984) was in many respects a complement to The Limits to Capital, and whose subsequent work on nature, scale and gentrification has been important and original. And, more recently, Erik Swyngedouw and Andrew Merrifield have, in different ways and at different levels, sought to bring historical-geographical materialism into the 1990s. This is not, of course, to say that these authors are all of a piece. Nor is it to suggest that their intellectual and political labours are limited to their geographical xiv commitment to a classical Marxist perspective. But it is to say that their works do form a corpus, one loosely - but nonetheless identifiably - structured. If this broader definition can be sustained, then my interest in historical-geographical materialism will not be too difficult to fathom. Even its detractors recognise that since its inception over two decades ago, it has been an approach to political-economy of remarkable theoretical and critical fecundity. It has made important, sometimes groundbreaking, contributions to our understanding of topics as seemingly different as the production of space and nature, urban growth, land rent, residential differentiation, urban politics, the local-global dialectic, urban consciousness, the spatial rhythms of global finance, the spatio-temporality of economic crises and the meaning and nature of socio-environmental justice. For this reason it seems to me to be more than worthy of sustained consideration. Of course, as I noted above, if one steps outside the disciplinary frame for a moment it may seem that I risk a certain parochialism in choosing to focus on a body of work - and of largely theoretical work to boot - specific to human geography. However, I would dissent from this view because I believe that the work of the authors I consider -particularly Harvey's but increasingly Smith's - has had an impact far beyond the disciplinary domain. Disciplinary boundaries cannot be wished away, of course, but I think that after a long period of 'importing' Marx into geography, Marxist geographers have begun to 'export' their insights and show to other critical theorists that 'geography matters' in Marx and should, therefore, matter to contemporary Marxists too. Indeed, for this reason I regard historical-geographical materialism not simply an adjunct to classical Marxism but, more emphatically, a constitutive part of it. And it is for this reason too that I began these introductory remarks with a consideration of the 'crisis of Marxism', not simply of Marxist geography, as if the latter, in its various forms, is some altogether less important and separate sphere. xv And yet today, perhaps in part because of this, historical-geographical materialism is arguably seen as a degenerating research programme. The frosty reception given to Harvey's The Condition of Postmodernity (1989) by several feminist critics seemed to mark something of a turning point for this version of Marxist geography within the critical geographical community. Since then, as the various 'posts-' have filtered into human geography during the 1990s, its fate has, not surprisingly, largely mirrored that of classical Marxism outside the discipline. For, in short, it too is seen is being too 'modern' - too immodest, too grand and too certain in what are, in any case, considered to be its outdated cognitive claims. Accordingly, human geography now has its own fair share of 'postmodern Marxisms', be they indebted to Resnick and Wolff (as, for example, in K. J. Gibson-Graham's work) or to Laclau and Mouffe by way of Hindess and Hirst (as, for example, in Stuart Corbridge's work). And, as a corollary, Sayer (1995) is right to note that radical political-economic theory in human geography has been attenuated, as a new generation of critical geographers choose to plough other intellectual furrows. It seems, then, that I have chosen a subject with little in the way of a future. However, it will be my contention that the classical canon which Harvey et al. have extended is far from exhausted. In fact, I will argue that it is possible to reveal in Marx's mature political-economy a problematic as captivating as any construed this century. A problematic which can help re-establish its relevance to the geographical Left; a problematic which can help reanimate historical-geographical materialism as a theoretical and critical enterprise. In other words, then, I am seeking to 'underlabour' on behalf of that which I assess: as a 'Marxist' on behalf of classical Marxism, as a 'Marxist geographer' on behalf of the historical-geographical materialism which has enriched Marx's original insights so creatively. My hope - to extend my earlier terminology - is to disclose a specifically 'late Marxist geography' for our times. These are ambitious claims, I realise, and I will try to make good on them. They are also claims likely to generate scepticism and hostility in equal measure. Scepticism from xvi those non-Marxist radicals within geography who think, with some justification it has to be said, that yet another reinterpretation of Marx can yield only diminishing theoretical returns. And hostility from those geographers on behalf of whose work I underlabour: for not only might they find my reading of Marx and their own work tendentious, but they may also find this effort of reinterpretation an unnecessary diversion. To both groups I can only ask that they suspend their disbelief until they have ventured where my arguments take them. My hope is that they find the critical tools I seek to fashion as productive as I do. This thesis has its origins in a series of papers published in geographic and non-geographic journals in recent years.81 mention this only to emphasise that what follows in not in any sense a simple stringing together and regurgitation of these earlier essays. While clearly indebted to their published forebears, the arguments presented here are substantially new and were, in many respects, only nascent within the journal papers they build upon. Aside from occasional passages which are reproduced more or less verbatim, this is, therefore, an original presentation. I should say also that, while I have tried to write in a lively and accessible manner (though in Part III confess that things necessarily get a little heavy), a good deal of what follows assumes a basic familiarity with Marx's later works and, of course, an at least basic acquaintance with the work of Harvey et al. NOTES "Of course, the modality of Marxism which, famously, lies on the cusp between what I am here calling 'modern' and 'postmodern' Marxism is that fashioned by Louis Althusser, who Crook (1991: 135) describes as 'the trojan horse of postmodernism' because he inadvertently pushed Marxism too far beyond its modern architectonics. I say more about Althusser in note 2. 2Jameson's (1990) Late Marxism also seeks to formulate a Marxism which can move beyond some ofthe old orthodoxies - but it does so in a way quite different from my own. Indeed, Jameson's work takes on a wider significance in relation to the kind of Marxism I argue for in this thesis because, like Perry Anderson, xvii his project has been to carry forward Louis Althusser's remarkable attempt to forge a Marxism which, as I observe in note 1 above, is neither wholly 'modern' nor wholly 'postmodern'. That Althusser's substantive theses ended in ruins is well-known, but his intent - to de-rigidify Marx's analysis without evacuating its 'scientificity' and explanatory power - remains exemplary and, in spirit at least, informs my own efforts here. More the pity then that most of the most original Marxist and post-Marxist work which has been inspired by Althusser this last twenty years (e.g. Resnick and Wolff 1987; Laclau and Mouffe 1985) has, as I noted on page ix, teetered on the edge of gutting Marxism of its core arguments and analytical substance. 3 An obvious and notable absence here is 'Second International Marxism', which posited 'laws of history', specifically of capitalism, deemed to operate with an iron logic, and which was largely associated with the former Eastern Bloc. Needless to say it represents a tendentious and vulgar rendition of Marx's critical inquiries. Callinicos rightly passes over it in silence since it can scarcely be considered a live option in any of the contemporary debates on Marxism. "Indeed, there has been a remarkable reawakening in Althusser's life and work of late. Among the more notable reassessments are Kaplan and Sprinker (1996) and Elliott (1994), building on the earlier appraisals of Benton (1984) and Elliott (1987). 5For some recent critical assessments of analytical Marxism, both in relation to Marx's original work and in its own right, see Carver and Thomas (1995), Hunt (1993), Marcus (1996) and Mayer (1994). 6In Anderson's definition Western Marxism includes structural Marxism too, which I have already picked out for separate attention. 7In other words, we need a synoptic account of contemporary Marxisms rather in the style for which Perry Anderson has become justly noted. His recent edited collection does this for Western Europe (Anderson 1996), but we await a wider survey of Marxisms today. 8Castree (1994, 1995a, 1995b, 1995c, 1996, 1997a, 1997b, 1998a, 1998b). xviii A NOTE ON TERMINOLOGY In this thesis I use several terms which take on quite specific meanings within the context of the arguments I develop. For this reason, and to avoid confusing readers, I think it useful to define these terms at the outset. Modern Marxism: this is a portmanteau, evaluative category which describes any modality of Marxism which critics consider to be guilty of the sins of cognitive universalism, foundationalism and authoritarianism. Postmodern Marxism: this is also a portmanteau, evaluative category which describes any modality of Marxism which critics consider to have moved beyond, or at least sought to move beyond, universalism, foundationalism and authoritarianism. Two leading contemporary examples are the Marxisms of Laclau and Mouffe (1985) and Resnick and Wolff (1987). In this thesis postmodern Marxism has a pejorative connotation since I consider the work the category designates to have evacuated, in the name of moving beyond supposed orthodoxies, Marx's project for a systematic, explanatory political-economy. Late Marxism: this is a general category which describes any modality of Marxism which is neither 'modern' nor 'postmodern' i.e. a Marxism which seeks to retain the coherence and explanatory power associated with Marx's original project for a critique of political-economy, yet which simultaneously seeks to address the aporias and rework the theoretical architectonics which makes modern Marxisms so objectionable without descending into the eclectic pluralisms of postmodern Marxisms. xix Marxism: a portmanteau, descriptive category which, in this thesis, can only be used in a strictly improper sense since it is really a catch-all category for a series of multiple and specific Marxisms, in the plural. Classical Marxism: this is a specific and descriptive, rather than evaluative, category. In this thesis it refers to a specific set of works by the later Marx (particularly the Grundrisse, Capital and Theories of Surplus Value) and to those epigones of Marx, including David Harvey, Neil Smith, Erik Swyngedouw and Andrew Merrified, who draw directly upon these works without seeking to blend, rework or dilute them with ideas drawn from other social and economic theories. Althusserian (or structural) Marxism: this is also a specific and descriptive, rather than evaluative, category. In this thesis it refers to the middle works of Louis Althusser and to the intellectual-political project inaugurated by those works. These works, despite their claims to be revealing the 'true' Marx, nonetheless fashioned a syncretic Marxism blending insights from Marx, Canguilhem, Bachelard and Spinoza (among others). In addition, Althusserian Marxism is indelibly present in a number of the recent attempts to move beyond Marxism, including postmodern Marxisms like those of Laclau and Mouffe (ibid.) and Resnick and Wolff (ibid.). Analytical (or rational choice Marxism): this is also a specific and descriptive, rather than evaluative, category. In this thesis it refers to works by authors like Jon Elster, John Roemer and Erik Wright, which blend Marx's arguments with insights and reasoning drawn from analytical philosophy and mainstream social science. Western Marxism: following Anderson (1976), this is a broad category which describes a set of post-classical continental European Marxisms - including Althusser's - which re-xx read and re-fashioned Marx using the tools of continental philosophy, social theory and aesthetics. It includes the work of thinkers like Lukacs and Benjamin, Adomo and Horkheimer, Sartre and Althusser and Colletti and Delia Volpe. Traditional Marxism: this is an evaluative rather than descriptive category and it refers to no one particular body of Marxian work. Rather, it designates a certain kind of reading of a particular body of Marxian work, a reading which is a more specific and concrete articulation of the general category 'modern Marxism'. Specifically, in this thesis traditional Marxism refers to any modality of Marxism which can be read as a critique of capitalism from the standpoint of labour and which instantiates a set of specific ontological and epistemological theses associated with this (detailed in chapter 2). Marxist Geography: a putative singular which describes geographical work underpinned by Marxian concepts. However, strictly speaking it is an improper category since its real referents are a set of different Marxist geographies in the plural, of which historical-geographical materialism (see below) is one. Historical-Geographical Materialism: this term was coined by David Harvey (1985a: xii) and is routinely associated with his project to spatialise Marx's political-economy. In this thesis, however, historical-geographical materialism takes on a broader and more specialised meaning. First, it is regarded as a research programme rather than the work of any one individual, a programme including Harvey along with Neil Smith, Erik Swyngedouw and Andrew Merrifield. Second, it is regarded as a geographically inflected form of classical Marxism insofar as its authors base their ideas upon direct appropriations and reworkings ofthe later Marx's writings. xxi T H E N A T U R E O F T H E P R O B L E M What they had regarded as a solution, [w]e consider ... but a problem. David Harvey (1973: 126, emphasis added) 1 CHAPTER ONE ENVISIONING CAPITALISM: MARXISM, GEOGRAPHY AND THE META-THEORETICAL IMPERATIVE ... we need be as suspicious of sheer novelty as we must be of leaden orthodoxy. Gregor McLennan (1989: 6) I. Introduction These are difficult times to be a Marxist in geography. Within a decade of its inception in the early 1970s, Marxist geography had established itself as perhaps the most significant critical approach within the discipline. It subjected topics as important and diverse as uneven development, urban growth, the geographical dynamics of manufacturing change, resource over-exploitation, Third World under-development, urban politics and social segregation to new and powerful interpretations. At the same time, it rejected the empiricist and positivist mind-set of mainstream human geography to show that geographical research is necessarily both explanatory and critical, rather than simply descriptive and value-free. As such, while not quite initiating a paradigm shift within the discipline, Marxist geography certainly legitimised the now common-place notion of a 'critical human geography'. This could be seen particularly clearly in the field of economic geography, once the preserve of neo-classical location theory. By the mid-1980s, Marxist and Marxisant authors such as Michael Dear, Costas Hadjimichalis, David Harvey, Doreen Massey, Richard Meegan, Richard Peet, Andrew Sayer, Allen Scott, Neil Smith, Ed Soja, Michael Storper, Richard Walker and Michael Webber had each in various ways articulated influential anatomisations of the capitalist space economy, ensuring that economic geography became less about 'economics' and more about 'economy' as a specific domain of social and spatial relations. A decade on, however, Marxism is no longer at the cutting edge of critical research in the discipline. For, of course, it did not take long to realise that the geographies of power 2 and inequality extended far beyond those generated by capital. For this reason, the late 1980s saw the rapid rise to prominence of several new critical geographies, notably those inspired by feminism and those concerned with 'race' and ethnicity. More recently, these approaches have been enriched and complicated by their engagements with post-modern, post-structural and post-colonial theories, which now enjoy considerable influence on the geographical Left. Together, these new critical geographies today co-exist with and supplement a no longer pre-eminent Marxist geography. But they go also much further than this: for they have been articulated in large part as a critique of their predecessor, a critique which has gathered pace and momentum throughout the 1990s. This has several dimensions, but if there is one charge which has become familiar through the sheer persistency with which it has been levelled, it is that Marxism is guilty of being a 'modern meta-theory': of being too essentialist, totalising and immodest in its cognitive claims and its normative imperatives. Deriving in part from Lyotard's (1984) assault on Enlightenment 'grand recits', this criticism today stands for much of what is taken to be wrong with Marxist geography by its erstwhile allies on the Left as they seek to escape foundationism in ontology, imperious rectitude in epistemology and a 'terroristic' politics in which non-class forms of otherness and difference are rudely shunted aside. The results of this critical assault have been two-fold. On the one hand, it has decisively shifted the centre of critical research away from traditional Marxian concerns towards a preoccupation with other social identities and relations. On the other hand, where Marxism is still drawn upon, it is almost invariably within 'post-Marxist' lexicons which seek to make it more flexible, open-ended and sensitive. These developments have, I think, been important and salutary. Because of them the geographical Left is today more ecumenical and vibrant than ever before, and new critical geographies - most recently gay and lesbian geographies - yearly enrich it. And yet, at the same time, I think something vital may have been lost here. Imean this in two 3 senses. Theoretically, the critique of Marxist geography is now so entrenched that we arguably risk relinquishing cognitive tools which may still, when appropriately reinterpreted, be of vital service to the contemporary geographical Left. As importantly, the broadening of the critical geographic community has been coincident with a declining thematic interest in radical political-economic analyses of what Sayer (1995: 1) calls the 'formal economy'. Accordingly, one looks in vain today for work with the ambition and imagination of - to take two excellent and well-known examples - Harvey's (1982) The Limits to Capital or Massey's (1984) Spatial Divisions of Labour. This is both regrettable and ironic. I say regrettable, because the turn away from Marxism comes at a time when the Right is arguably enjoying an unprecedented economic and political hegemony in buisness and government worldwide. And I say ironic, because with the collapse of communism and the continued integration of less developed countries within an international political-economy, we can for the first time proclaim with some confidence that we now live in what Marx called a 'world market'. The geographical consequences of this have been profound. The last few years have been marked by uneven and dynamic patternings of capital investment and disinvestment resulting in the formation of a new and highly volatile mosaic of industrial activity at the regional, national and international scales. Just as new spaces of accumulation have arisen virtually overnight, so previously prosperous urban and regional production complexes have collapsed with equal rapidity, leaving a legacy of human and environmental devastation and suffering. At its most extreme, this patterning of uneven development has entailed the boycotting of certain regions altogether, most notably sub-Saharan Africa where, as Smith (1990: 165) puts it, "the holy texts of progress have wrought nothing less than a swathe of satanic geographies". Living at the end of the twentieth century, Marx and Engels' image of solidity melting into air has perhaps never seemed more appropriate. And yet, if one is to believe the now commonplace mantras issuing from buisness interests and mainstream politicians, the economy is no more than a collection of monadic 4 individuals and communities, each sovereignly and equally responsible for their own destiny. This rhetoric, which has its origins in a resurgent neo-classical economics in many universities and government think-tanks, sits well with the neo-conservative climate ofthe times. But missing here are questions of class and social relations; and missing too are questions of power and inequality. In short, there is no sense here of economy as at once 'social economy' - that is, a whole "way of life, founded in production" (Peet and Thrift 1989: 3) - and 'political-economy' - that is, a way of life founded on the production and uneven distribution of an economic surplus.1 For this reason, it seems to me absolutely vital to 'reclaim the reality' of the space-economy in the interests of progressive change.2 Put differently, I would argue that we (still) need to pursue a contemporary project of envisioning capitalism: that is, a project of making critically visible the political-economic relations we collectively constitute, are embedded in and subject to on an increasingly worldwide scale. Without such a critical project we risk allowing the Right's 'market Utopians' (as Maurice Glasman [1996] calls them) to obscure the violent realities offin-de-millenium political-economic life behind the visage of a faery-tale world of happy producers and satisfied consumers. However, any such project immediately faces a dilemma. On the one hand, there exists a body of theory upon which we can draw which seeks to envision the space economy in exactly the way I am arguing for: Marxian political-economy. On the other hand, though, the critique of Marxism elaborated in geography these last few years means that we cannot simply 'return' to this corpus in a straightforward act of recuperation. This may seem immediately to point us in the direction of the 'postmodern' Marxist geographies to which I have referred for theoretical aid and assistance. However, I have two difficulties with such a rush to the 'post'. First, as I will explain shortly, while I fully sympathise with the intent of these attempts to move beyond perceived Marxist orthodoxies, I am not convinced they always offer a systematic and coherent alternative account of capitalist economic relations. It is if the critique of Marxism per se is more important than 5 developing an alternative account of capitalism. Second, while many of the criticisms levelled against Marxist geography on the geographical Left are surely on the mark, it is my belief that the critics have not always correctly identified the subject of their censure. That is, their critique of that which they seek to supercede - Marxist geography - has not always been as precise and discriminating as I think it should rightly be. As a result, they run the risk of dismissing Marxism before all the evidence is in. In light of all this, it is my intention in this thesis to reclaim a specific body of Marxist geographical work for a contemporary project of envisioning capitalism. This cannot, as I have indicated, be in any sense a simple reclamation, that is to say, a stubborn restatment of already established Marxian geographical 'truths'. But nor, as I have also indicated, do I think it necessary or necessarily productive to take a detour through the various post- 'isms' in order to read Marxism anew. Instead, I think it is possible to find a 'third way' for Marxist geography. By this I mean one which can address and move beyond some of the very real limitations identified by the critics, which can thereby remain theoretically open, modest and reflexive, but which can also retain the coherence and explanatory power necessary for Marxian theory to function effectively as a critique of the capitalist space economy. If such a 'both/and' geographical Marxism can be located then it holds out the promise of re-engaging Marxism in geography with the discipline's wider critical community and of uniting both against the free market orthodoxies that obscure the satanic geographies endemic to capitalism. These are bold and ambitious aims and assertions, and they indicate a number of tasks to be undertaken simultaneously. They implicate me in working on behalf of Marxian political-economy in geography and yet in some respects against it; in dissenting from some of the plenary dismissals of Marxism by the wider geographical Left while concurring with its imperative to take complexity, otherness and difference seriously; and, in all cases, in working against neo-conservative representations which reduce economic life to a bloodless flow of goods and services to be measured by supply and demand 6 curves. This does not, I think, make me a theoretical schizophrenic. But it does require me to show that all these roles can be occupied in a coherent and consistent manner, and I hope to do that in some considerable detail in the chapters which follow. If these comments indicate my basic purpose, they only apparently identify the exact body of work to be returned to and reclaimed. I say this because - although the critics have not always made this explicit - 'Marxist geography' is not a singular entity, but instead a heterogenous set of overlapping discourses which put Marx's political-economy to work in different ways (Castree 1994). This is a extremely important point, because any proper consideration - sympathetic or critical - of Marxist geographical work cannot proceed on the basis of such abstractions (Watts 1988). Instead, it must be based upon a careful specification of the theoretical terrain(s) to be surveyed. There are today several extant Marxist geographies rather than simply one, which vary according to their interpretation of Marx, their explanatory reach and their influence on the wider geographical (and non-geographical) community. With this in mind, the focus of my reclamatory energies in the following pages is a modality of Marxist geography which I am not alone in regarding as distinctive, innovative and theoretically powerful: what David Harvey (1985a: xii), its chief advocate and practitioner, has famously called 'historical-geographical materialism'. It is an indication of the misunderstandings here that this term is routinely associated with Harvey's work alone, or else is taken to be a synonym for 'Marxist geography' tout court. However, I think it is more accurate and productive to view historical-geographical materialism as a research programme, consisting of the work of several individuals - notably Neil Smith, Erik Swyngedouw and Andrew Merrifield - of which Harvey is but one. A basic, but accurate, description is that it is a research programme distinguished by an attempt to creatively extend Marx's original political-economic analyses in a geographical direction. In other words, it is not a Marxism read geographically through the texts of any of Marx's post-classical reintepreters. It is, of 7 course, difficult, if not impossible, to draw any paradigmatic boundaries in these situations. To those names I have mentioned there are others I could add (for instance, Dick Walker and Richard Peet) and there are key individuals, notably Doreen Massey, whose work I do not consider in this thesis. But here the Lakatosian notion of a research programme is useful, because it suggests a field of scholarship at once internally diverse and fuzzy on the margins, yet at some level possessed of a determinate core. It will be my suggestion that the efforts of the few authors I have named possess a structured coherence that legitimately allows one to label them a research programme in just this sense. Seen thus (and I will, of course, need to fill out this bare bones definition), it is a research programme of uncommon breadth and ambition which has made a very significant contribution to our understanding of capitalism's various geographies. In the work of David Harvey and Neil Smith, and more recently that of figures like Swyngedouw and Merrifield, historical-geographical materialism has addressed itself to such vital questions as the role of space in capital accumulation, the dynamics of nature's production, the necessity for capital to urbanise, the space-place dialectic, the nature of land rent, the spatio-temporality of economic crises and the meaning of social and environmental justice, to name but a few. Harvey, of course, has been the most significant presence here. His remarkable The Limits to Capital (1982), in many ways the charter document of the movement, offered no less than a wholesale, spatialised reconstruction of Marx's critique of political-economy, and has been succeeded by several important books and essays.3 But Neil Smith's ambitious Uneven Development (1984), and his subsequent work on the social production of nature, the production scale, and on rent and gentrification, the work of Swyngedouw on money, on industrial restructuring, on space as a force of production, and on the political-ecology of water, and that of Merrifield on place have all been very important and influential too.4 In short, then, in its distinctive adherence to and extension of the classical Marxist texts, historical-geographical materialism has established for itself a particular and important place within the wider Marxist geography movement these last 8 two decades. And in so doing, it has powerfully illuminated previously hidden geographies of capital and class. If this counts as one reason why I choose to focus on it here, another is the wider influence it has exerted, firstly within geography, and secondly outside the discipline. Whether its specificity is recognised or not, it has arguably been the most influential of all existing Marxist geographies on the wider geographical community. Harvey's work, in particular, has exerted an enormous influence in urban and economic geography, and, whether in appreciation (e.g. Gregory 1991) or sharp criticism (e.g. Massey 1991), has been recognised as an corpus of signal importance within contemporary critical human geography. But in recent years, the insights of Harvey et al. have also begun to be registered outside geography too. In his history of Marxist geography and Western critical theory, Soja (1989: 56) has suggested a two-stage, two-way relation between the two fields. Where geographers imported Marxism into their discipline during the 1970s and 80s, they have since, he argues, reversed this parasitic relation in a "provocative inversion" wherein Marxist geographers are now showing other Marxists that 'geography matters' in any political-economic analysis. While I am not so sanguine as Soja, I think he does have a point, and it is pleasing to see how many other critical theorists today take a serious interest in Marxist geographical work by Harvey, Smith and others. And yet, this double importance notwithstanding, historical-geographical materialism is generally seen as a research programme which is now in decline. As critics have assimilated its failings to those of a generalised 'Marxism' and 'Marxist geography', its influence has waned and interest in it diminished. To my mind, Harvey's recent apologia - Justice, Nature and the Geography of Difference (1996) - has exacerbated rather than ameliorated this state of affairs. His most ambitious book by far, in it he tries to defend historical-geographical materialism from its detractors by showing how it can, seemingly, extend to explaining most of the major concerns of contemporary critical theorists - including those of social difference, environmental abuse and socio-9 environmental degradation. This is not the place to review the book in any detail (see Castree 1997c, 1998a; Clark 1998; Eagleton 1997), but it seems to me that Harvey has here squandered the opportunity to engage his Marxism in an exciting and progressive way with his critics. Like him, I too believe that the theoretical resources of historical-geographical materialism are far from exhausted; but unlike him, I think a substantial effort of immanent critique and reinterpretation of its most central concepts is required to renew this body of thought in a productive fashion. In the remainder of this chapter I want to expand on these introductory comments in a step by step fashion in order to lay the groundwork for the chapters which follow. This will involve me in a qualified defense of political-economy which is both meta-theoretical and systematic, a further consideration of the importance of 'envisioning capitalism', a further specification of the field to be critically reappraised and a discussion of the stakes involved in 'reading theory' and in 'underlabouring'. But as a way into these important issues I begin by assessing the accuracy and effects of the prevailing criticisms of Marxism in human geography. II. The new orthodoxy? Genuine refutation must penetrate the power ofthe opponent and meet him (sic) on the ground of his strength; the case is not won by attacking him somewhere else and defeating him where he is not. Theodore Adorno (1982: 5) As I noted above, the recent blossoming of new critical geographies has been achieved in large part through a critique of the Marxist geography which preceeded them. This has several dimensions, but the most serious and persistent charge is that Marxism succumbs to 'meta-theoretical' impulses. This has three aspects which are seen to be particularly serious and intrusive: (a) foundationalism: foundationalism designates discourses which appeal to archai or irreducible first principles of social life which can ground and explain all others. In its 10 moderate form foundationalism entails prioritising certain generative processes within a social formation for special attention; in its more extreme form it entails what Resnick and Wolff (1987: 2) call 'essentialism1, that is the reduction of social life to all-important 'essential processes'. Betraying the Enlightenment origins of the theory upon which it builds, Marxist geography has been seen as particularly susceptible to both forms of foundationalism. Thus Julie Graham (1990), in her essay on 'Theory and essentialism in Marxist geography', has argued that Marxist geography erroneously posits 'production' or the 'economy' as ultimately determinant, be it in the first instance or the last. Likewise, drawing on Laclau and Mouffe, Stuart Corbridge (1989) has argued forcefully for a 'post-Marxist geography' which can put capital and class in their proper place. And, in a not dissimilar contribution, Trevor Barnes (1989a) has turned to Rorty's critique of the search for guarantees for knowledge in order to upset Marxist geography's essentialisation of the socio-spatial sphere. The suggestion, then, in each of these cases, is that Marxist geography urgently needs to shift into a 'post-foundationalist' register where production and economy are no longer deemed to be ontologically pre-eminent but merely one of several equally important social spheres. (b) universalism: if Marxist geography has been accused of foundationalism, then a corollary complaint is that it has sought to apply its foundational insights to all times and all places. This universalism is not secured simply by an act of theoretical imposition, but by a presumption that "behind all our seemingly ephemeral and evanescent experiences ... [is] a hidden order" awaiting discovery (Barnes 1996: 7). In this sense Marxism is regarded as a 'totalising' theory, one in which, to use Habermas's (1991a: 332) felicitous words, "the transcendent moment of universality bursts every provinciality asunder". Accordingly, Marxist geography has been reprimanded for claiming to be the "only perspective" (Gould 1988), for claiming to take "the view from nowhere" (Deutsche 1991), and for running roughshod over the complexities which differentiate social formations (Ley 1989). In part, 11 this cognitive universalisation is seen as being achieved through theoretical abstraction, the generality of theory refusing to engage with and be complicated by the messy particularities of concrete conjuncture. But more pointedly, several critics have argued that universalisation is in fact implicity masculinist and Eurocentric, deriving from an Enlightenment tradition which depended upon patriarchal authority and Western political-economic supremacy (Gregory 1991; Massey 1991; Rose 1989). It has thus become axiomatic within critical geography to argue that theory must be situated in such a way that the risks of making it 'travel' can be closely examined. As a result, meta-theoiy now has a bad name in leftist geographical circles, and re-assertions of it - as in Harvey's Justice, Nature and the Politics of Difference - not surprisingly generate intense suspicion. (c) authoritarianism: a third perceived failing of Marxist geography, which follows directly from the first two, is that it is authoritarian. Apparently grounded in socio-ontological certainties, Marxist geography's exorbitant meta-theoretical claims have been articulated with a force and authority which is no longer seen as realistic or desirable. Like other modern intellectuals, Marxists, within geography and without, have routinely regarded themselves as what Bauman (1987) calls 'legislators': that is, transcending their own situated circumstances, they seek to pronounce authoritatively on the course of society in general. Against this, critics have argued that Marxists should adopt the more modest role of 'interpreters', of intellectuals who adopt a reflexive stance with regard to their own knowledge productions and who evince a modesty in their theoretical propositions. This argument has been made with particular power by Rosalyn Deutsche (1991) and Megan Morris (1992), who both rightly object to the trenchant claims Harvey makes in The Condition of Postmodernity about the apparently unidirectional relations between economic and cultural production. But it has since been reinforced by others, who now firmly believe that theory in a minor key is the only viable way forward for any contemporary critical geography, Marxist or otherwise (Katz 1997). 12 Taken together, these now familiar accusations of foundationalism, univeralism and authoritarianism amount to a powerful critical armoury with which to attack Marxism's and Marxist geography's claims to know. I have little doubt that a good deal of Marxist geographical work has been arrogant and arrogating in the way the critics suggest, and that attempts to move beyond this 'modern' mode of working are appropriate and timely. Indeed, as I noted above, the two-fold effects of this assault - the creation of a space in which other forms of critical human geography can flourish and achieve legitimacy, and the reformulation of Marxist geography within various post-modern and 'post-Marxist' frameworks - have been on the whole immensely positive. And yet I want to register several reservations, because it seems to me that the critique of Marxism in geography has in some respects also been detrimental. My first reservation relates to the supposed accuracy and exhaustiveness of the triple-headed critique of Marxist geographical theory. For what is striking about much of the criticism which I have cited is that it is directed against 'Marxist geography', a category I have already objected to as being incohate. A peculiar thing happens here, wherein the performativity of the name calls forth the thing it ostensibly represents. For the very use of the portmanteau term 'Marxist geography', with a capital M and in the singular, actively obscures difference and brings together different Marxisms within a more general interpretive horizon where the sins of the undifferentiated parts are transferred onto the putative whole, and the whole in turn used the question the worthiness of the parts. To be sure, this synechdochal maneouvre is often unconscious and unintended, but it nonetheless functions to hinder the kind of careful immanent critique which I think is vital if Marxist geographies are to be accurately appraised. This is not to say that all the works listed above take issue with Marxist geographies somewhere else and defeat them where they are not. On the contrary, Deutsche's (1991) 'Boys town' is an excellent example of a critique which focusses in-depth on a particular example of Marxist work: Harvey's The Condition. And yet even here 13 I think there are problems: for if critique can be too undiscriminating it can also be too specific and single-minded. I say this because Deutsche's impassioned rejection of Harvey's meta-narrative claims focusses on one - in many respects unusual - work, without taking into account his wider theoretical corpus. Moreover, in her determination to expose Harvey's sins, her reading of his Marxism becomes as one-sided and absolute as his of postmodern culture.5 In this respect, Gregory's (1991; 1994 ch. 6) interventions strike me as among the few examples of a critical reading where the multiple strands of Harvey's brand of Marxism are brought out and subjected to careful comparative scrutiny. If, then, I find the critics not altogether certain as to the object of their animus, I also worry about the exhaustiveness of their criticisms. Words can have something lethal about them, and here it is arguable that the charges of foundationalism, universalism and authoritarianism have solidified into a 'new orthodoxy' about Marxism in geography which forecloses the field of debate. Ironically, this squeezing out of other possible interpretations of Marxist geographical work sits uneasily with the spirit of the times. After all, do not so many critics of Marxism insist that our contemporary 'structure of feeling' is characterised by a respect for and sensitivity to polyphony, diversity and difference? If this is so, and if, as I believe, any substantial and varied corpus of theory invites multiple legitimate interpretations, not plenary prognostications, then surely - to play on Althusser's memorable words - the lonely hour of the last word about Marxism should never come. These comments bring me to my second complaint: the declining interest in analyses of the capitalist space economy. Of course, Marxism was not the only brand of political-economy to undertake such an analysis in geography (see, for example, Sheppard and Barnes' [1990] excellent survey and geographical extension of analytical political-economy). And, in all other respects, the shift of geographers' research interest towards domains of social life not traditionally considered 'capitalist' or 'economic' is important and exemplary.6 And yet capitalist social relations and forms still persist - pervasively and intrusively - and therefore an effort of theoretical labour is still required to fashion the 14 concepts and explanations to disclose the logic and direction of those social ties. As I hope to show, despite existing critiques, Marxist political-economy still has a great deal to offer here. This is true in several respects, but one which I regard as particularly important is precisely the meta-theoretical impulse critics have found so objectionable. This is not as perverse as it seems, because while I concur with the view that some meta-theoretical claims to know are exorbitant I also believe that some form of meta-theoretical political-economy is still necessary, relevant and defensible. My third, and final concern, relates to the the work of those who have sought retain some of Marxism's insights but in non-meta-theoretical ways inspired by the gamut of 'post' -isms that now pepper the intellectual landscape. Two recent and particularly accomplished examples of such work immediately come to mind, both written by authors who in the past have been sympathetic to Marxism. The first is K-J Gibson-Graham's (1996) The End of Capitalism (As We Knew It): A Feminist Critique of Political Economy. Imaginative, ambitious and well crafted, the book's central claim is that theoretical political-economy has called forth a Leviathan - 'capitalism' - which is too big, too totalising, too disabling. By eliding their representations of the 'real' with the real itself, Gibson-Graham argues that discourses like Marxism have colonised the cognitive space which would otherwise be available to rethink 'capitalism' as incomplete, messy and fractured. Accordingly, The End of Capitalism (As We Knew It) seeks to bring capitalism down to size and explore the interstices where class, gender and sexuality mutually constitute one another. The second example is Trevor Barnes' (1996) Logics of Dislocation: Models, Metaphors and Meanings of Economic Space. Echoing Gibson-Graham's aversion to theories which posit ontological monoliths like 'capitalism', Barnes objects to the Enlightenment narratives through which much of economic geography has prosecuted its case. Instead, he advocates 'post-prefixed' theories of the space economy which abjure absolute certainty, cognitive exhaustiveness and ontological groundedness in favour of modesty, complexity and contingency. Drawing inspiration from such diverse 15 post-foundational approaches as the sociology of scientific knowledge, Derridean deconstruction and Rortian pragmatism, Barnes turns to the likes of Pierro Sraffa, Harold Innis and Fred Luckerman for detailed exemplifications of the post-Enlightenment mode of theorising he seeks. Both these books are innovative and important and will deservedly become the subject of much debate and discussion. They unravel the systematicities of geographical political-economy, not in an act of wanton destruction, but in a reconstructive attempt to reveal geographical worlds marginalised and obscured by these very systematicities. And yet, despite these attributes, two things strike me about these reconstructive endeavours. First, both leave Marxism so far behind that they arguably gut it of any coherence and explanatory power it may once have had.7 Second, in place of such 'old style narratives' both books offer suggestive insights for a post-prefixed political-economy, but fail to offer any compelling new theory of the capitalist space economy. I appreciate, of course, that this may be the point. But this does not sidestep the fact that some sort of explanatory account of contemporary economic life is both necessary and important. In this regard, Andrew Sayer's (1995) Radical Political Economy: A Critique - in many ways a complement to the two books cited - is to my mind a more productive attempt to think geographical political-economy anew (see Bassett 1996a). With these three reservations in mind, I want now to offer some general comments on why I think there is still a place for a Marxian political-economy which is specifically meta-theoretical and systematic. III. Marxism and the meta-theoretical imperative ... grand views are as necessary as ever ... Andrew Sayer (1993: 336) In my opinion one of the most important effects of Marxism's entry into human geography during the 1970s was its vigorous advocacy and defense of theoretical endeavour and, specifically, of critical theorising. Where spatial science rested on a positivism that 16 supposedly allowed the 'facts' to speak for themselves, Marxists argued for the ineluctability of theory and of explicit theory building. Though now dated in several rather obvious respects, Harvey's (1973) 'Revolutionary and counter-revolutionary theory in geography and the problem of ghetto formation' is still perhaps the most lucid account of why any critical human geography simply has to be theoretical. In the first place, Harvey argued, empiricism is simply not an option. As his critique of Alonso and Muth's urban land-use theories showed, the facts never come to us unmediated by cognitive assumptions and presuppositions. This is why Harvey insisted that theoretical categories be made explicit, rather than left implicit, so that they could be subject to debate and scrutiny. But more emphatically, Harvey also showed that theory 'matters'. He meant this not just in the simple sense that the categories we rely on make a difference to how we see the world, but in the more important sense that serious theoretical work entails the construction of an elaborated conceptual apparatus that actively tries to disclose and explain the world. In this second sense, theoretical knowledge can become 'consequential' in something like Culler's meaning of the term: "The works we allude to as 'theory' are those that have ... the power to make the strange familiar and to make readers conceive of their own thinking, behaviour and institutions in new ways" (Culler 1983: 9; see also Culler 1997). Social Justice and the City was - as Harvey intended it to be - consequential in just this way, and since its publication he and many other Marxist geographers have remained dedicated to rigorous and explicit theory construction. This legacy, happily, lives on. Despite the critique of Marxism in geography, the contemporary left of the discipline has retained, enriched and complicated this commitment to theory. Whether feminist or queer, post-modem or post-colonial, theory has become a medium to be worked with in critical human geography, openly and - as in the case of Gibson-Graham and Barnes - often very creatively. The main difference is that unlike Marxism these other critical geographers abjure meta-theory and prefer, instead, to openly and honestly 'situate' their claims and voice them with care and reflexivity. 17 So far so good. Except that to my mind meta-theory does not necessarily equate with stern foundationalism, arrogant universalism and unthinking authoritarianism. In other words, it seems to me that there are other possibilites for Marxian meta-theoretical inquiry. I quite agree that theory must be situated. But I dissent from the vulgar version of this argument in which theory can never exceed the horizons of its own particular site of production. Instead, handled sensitively, what I would call the meta-theoretical imperative is cognitively and politically vital in three respects that define its field of operation. In the first place, by showing how particular and local are often constitutively tied to general or even global processes and relations it transcends the ineluctably specific conditions of its construction to actively reveal worlds we otherwise could not see. Second, as a corollary of this cognitive mapping it makes claims on both ourselves and those "distant strangers" (Corbridge 1993) to whom we are related by virtue of these general processes. But for meta-theory to become truly 'meta' - and this is the third facet defining its field of operation - it must also escape its sites of production and be put in motion as a global formulation. This is far from easy, because in practice it requires enormous efforts of communication and translation between different actors in different locations - not the compulsive imposition of supposedly singular truths. When these points are added together it seems possible to draw two conclusions. First, meta-theory has a disclosive capacity which is relinquished only at a considerable cost. But, second, it is a capacity which must necessarily be handled sensitively and responsibly. This, I hope to show in due course, must be possible, since the alternative is precisely the kind of 'bad' meta-theoretical practices for which some Marxists have been rightly reprimanded. If I thus think it necessary to retain meta-theory in some form, I also think it equally important to do so systematically. I have alluded to this already. What I mean here is the development of political-economic theory which combines coherence with genuine explanatory power. As I said, I think that some of the 'post-Marxist' writing in human 18 geography has perhaps attenuated some ofthe better insights of Marxist geographical work in the process of criticising some of its less desirable elements. But more emphatically, the arguments of Gibson-Graham, Barnes and others also indicate a certain aversion to systematic political-economy because systematicity is equated with closure and inflexibility. As Bruce Roberts (1996: 193) puts it, "Systematicity in theory has come to be perceived ... as a flaw in itself. However, again I would resist such an equation and reduction because I will show that systematic Marxist political-economic theory can remain open-ended and flexible. Indeed, it must be if it is to combine a commitment to explaining the world with an equally important commitment to recognising the limits of its own categorial claims. IV. Envisioning the economy The incapacity to envision the economy can play into the hands of a reactionary ... [particularism] that thrives precisely on the condition of blindness to the determinates of contemporary social life. Susan Buck-Morss (1995:466) The meaning and pertinence of these reflections on systematic meta-theoretical political-economy (a phraseological mouthful if ever there was one!) can be rendered clearer and given a sharper edge if I elaborate further on the project of'envisioning capitalism'. According to the capsule description I offered above, this is a project of making critically visible the political-economic relations we8 collectively constitute, are embedded in and subject to on an increasingly worldwide scale. I want to stress the words 'critically visible' here because the term envisioning captures this sense oi making seen the economic connections which might otherwise remain opaque. This is not quite as banal as it might appear. The faculty of vision within theory has become fiercely contested of late. On the one side, as part of the general assault on Marxism a number of critics have demonstrated the complicity between meta-theory and certain modalities of vision. Within geography it is Harvey (1989a), once again, who has come under particular fire here. His assertive ocularcentrism in The Condition, so Rose (1989) and Morris (1992) claim, becomes a 19 faculty of cognitive exclusion in which Marxism adopts the Archimedean conceit of the perfect, total view of Society. However, on the other side, as this critique has proceeded the discursive space vacated by an embattled Marxism, within geography and without, has been increasingly occupied by neo-classical and liberal economic theories which, as I have said, refuse to focalise capitalism's relations of power and inequality. In light of this, I think that critical theorists simply must deploy the connective imperative between theory and seeing if they are to reclaim political-economic reality from the Right. But in light of the critique of vision, the key point is to do so in a way that undoes the Archimedean conceits of'modern' theory (cf. Rose 1997: 308-12). A lucid account of the stakes involved in envisioning the economy and the form such an envisioning might today take has recently been offered by political theorist Susan Buck-Morss. In 'Envisioning capital: political-economy on display' (1995), she examines the way discourses of political-economy since the eighteenth century have made the economy visible, particularly those presently ascendent versions of what she calls "market theory". Her critical point is that their "minimalist vision" consigns to invisibility "the web of social interdependence produced by economic activity". Indeed, by passing off the apparently atomic activities of seemingly asocial individuals as the 'truth' of economic life, market theory encourages a reactionary localism to re-enchant the 'empty' space of society, a localism presently seen in parts of Eastern Europe newly won to capital. In its place, therefore, she recommends a "philosophical, critical vision of the social body as it is produced by the global economy ..." to fulfill a "visionary need" to see "the social whole" {ibid. 465-66). Despite its political urgency, this plea appears at first blush to repeat the rather unreflexive ocularcentrism critics detect in the work of Harvey and other Marxist political-economists. But Buck-Morss's argument turns out to be more subtle than this. To begin with, 'the economy' of which she speaks does not pre-exist political-economy and thus await 'discovery'. Rather, "the discovery of the economy (during the eighteenth century) 20 was also its invention" (ibid. 439, 440). The economy, in other words, emerges as what John Rajchman (1991: 81) calls a "space of constructed visibility". Explaining this, Buck-Morss continues because the economy is not found as an empirical object among other worldly things, in order for it to be "seen" by the human perceptual apparatus it has to undergo a process, crucial for science, of representational mapping. This is a doubling, but with a difference; the map shifts the point of view so that viewers can see the whole as j/from the outside, in a way that allows them, from the inside, to find their bearings (ibid, emphasis added). Here, then, 'the economy', as Louis Althusser (Althusser & Balibar 1970: 101) once said in his theorisation of a decentred totality, "is precisely not expressed at all", for the concept "like every concept, is never immediately 'given', never legible in visible reality: like every concept this concept must be produced, constructed by the analyst". This is a vital point which is easily misunderstood. The suggestion is not that 'the economy' is simply a theoretical construct: for Buck-Morss capitalist relations are quite real and global in reach. Rather, her claim is that the theoretical labour involved in envisioning these relations has an essential world-disclosing function in which theory's performativity and consequentiality becomes plainly evident. To be sure, there is always the risk that political-economy can conjure up a representational monster - hence, for example, Gibson-Graham's concern. But without the kind of cognitive mapping Buck-Morss argues for there is equally the risk of abrogating altogether the responsibility to represent wider sets of social relations within which the theorist and his/her audience are mutually embedded. In this light, Buck-Morss's concluding call for a renewal of theoretical attempts to 'see' economy as "a social whole" thus becomes a subtle strategic and ironic intervention in the ongoing struggle to alter the workings of world-changing social systems such as that Marx called 'capitalism': we see 'the whole' as if from the 'outside', in a way that allows us, from the 'inside', to find our bearings. Derek Gregory (1994: 345) suggests the possible productivity of this reflexive strategy when he insists that, "there is no need to convert the 21 critique ofthe gaze into a recoil from vision". In this regard, Buck-Morss's most original point can be rendered thus: the struggle to conceptually 'envision' the economy is necessary to lay claim to that 'reality' even though it can never offer an entirely adequate, decided vision of that 'reality'. Aware of the dangers of disclaiming a view from somewhere, Buck-Morss nonetheless wishes to strategically reappropriate vision for political-economic critique. Theory here dares to make meta-theoretical propositions; but at the same time it never loses sight of the particularity and partiality of its vision. In Haraway's (1992: 295) words, it becomes "a little si(gh)ting device in a long line of such craft tools" - but a potentially powerful one nonetheless. V. Historical-geographical materialism and immanent critique A criticism that still struggles with its object remains dogmatic. Karl Marx (1975: 158) It remains, of course, to show how in practice Marxist geographical work can be reclaimed for such a reflexive project of envisioning the space economy. But this depends upon the prior task of identifying clearly the modality of Marxist geography to be reclaimed. And it is here that I want to take further my concerns about the critics' imprecision and my claim that 'historical-geographical materialism' be regarded as a distinct research programme. If critique is to be effective it must be immanent to its object. Yet, as I suggested, there has often been a striking lack of precision in the current critique of Marxist geography. This continues a pattern laid down by Duncan and Ley (1982), in what was the first major critical salvo against Marxism in the discipline. In 'Structural Marxism and human geography: a critique1, Duncan and Ley managed to conflate the work of a whole raft of Marxist geographers - Harvey included - with the then fashionable Althusserian Marxism dominant in Anglophone circles.9 This is not to say that their critique was wholly inaccurate - on the contrary, several important points were made - but that much of it stuck only be default. Likewise, it seems to me that the current critique of Marxism has often been on the mark in a similarly indiscriminate way. 22 As a way, then, of approaching and situating the object of my reclamatory efforts -'historical-geographical materialism' - let me offer a rough sketch of extant Marxist geographies (see also Castree 1994 and Reynolds 1994). I can do this in both a 'negative' and a 'positive' sense. The former entails indicating which Marxisms have not impacted upon human geography, even though some believe they have. Of these two stand out: Althusserian Marxism and the broader movement Anderson (1976) calls 'Western Marxism'. Contra Duncan and Ley, Althusserian Marxism - that novel blend of Marx, Spinoza, Canguilhem and Bachelard - has had virtually no substantive and systematic impact upon Marxist geography (unless sociologist Manuel Castells' The Urban Question [1977] is included). Instead, as I will suggest momentarily, Althusserian Marxism has had a less explicit and much more subterranean effect. Likewise, Western Marxism - that heterogenous continental reworking of Marx from Lukacs in the 1920s to Colletti in the 1970s - has, contrary to Soja's (1989) over-blown claims, barely insinuated itself into geographical research. Perhaps the only (and signal) exception here is work of Henri Lefebvre (see Gregory 1995: ch. 6). The Marxisms which have insinuated themselves fall under five, by no means exclusive, categories (in no particular order): (a) 'world systems' Marxism: drawing upon Wallerstein's magisterial reworking of Marx's corpus, Peter Taylor is perhaps the most forceful proponent of 'world systems theory' as an approach to economic and political geography. (b) regulationist Marxisms: the 'regulation theory' proposed by Algietta, Lipietz and others has had an enormous impact on human geography, particularly in terms of the Fordism/post-Fordism debate. Developed very much in the shadow of Althusser's thinking, it is today a far from unified post-Althusserian approach blended with other intellectual influences. However, several Marxists in geography - including John Lovering as well as 23 Harvey and Swyngedouw - have drawn upon it and brought out its implicit geographical dimensions. (c) realist Marxisms: at an altogether more abstract level, several geographers have approached Marxism through the lenses of Bhaskar's critical realist philosophy (and here once again we see the subterranean influence of Althusser, from whom Bhaskar has drawn so much). Seeking to move beyond what are perceived to be the limitations of orthodox Marxism, these authors - including Dick Walker and Andrew Sayer - expound a more open Marxism in which capital is merely one set of 'structures' and 'mechanisms'. Although it is diffcult to categorise her Marxism, Doreen Massey's middle work to a considerable extent fits into the realist Marxist mode.10 (d) cultural Marxisms: this is a broad and thematic category which is associated with no one particular brand of Marxist geography but which includes those geographical Marxisms inspired less by Marxian social science and more by the humanities and more culturalist Marxisms. Both Denis Cosgrove and Derek Gregory (in some of his earlier work: e.g. Gregory 1982) fit into this broad category, with the work of Raymond Williams and E. P. Thompson, respectively, influencing the work of each author. (e) analytical Marxism: this approach, the newest in Marxist geography, has been advocated by Barnes (1989b), among others, as a possible route beyond the limitations of older forms of Marxist geographical work. Drawing upon Elster, Roemer and other rational choice thinkers, it is committed to reworking Marxism with the tools of analytical logic. There are other more middling Marxist geographies I could mention, but the point of this rough map is not that it is exhaustive but that it throws into relief the specificity of historical-geographical materialism. For, in relation to these other Marxist geographies, 24 what is notable is that it is a fundamentally classical version of Marxism. In other words, it is a Marxism based on a direct reading and appropriation of Marx's original critique of political-economy rather than a 'reconstructed' Marxism read through the eyes of any of Marx's twentieth century post-classical epigones. This is not to say that other influences are altogether absent - on the contrary, Harvey, Smith, Swyngedouw and Merrifield have all drawn in different ways upon the work of Bertell Oilman and Henri Lefebvre among others. But it is to say that these other influences are used to enable and extend - not deconstruct - what is taken to be a more or less 'faithful' reading of Marx. This is highly unusual and not only, as I indicated in the Preface, within geography. For this reason, Harvey (1987: 369; 1986) has explicitly contrasted historical-geographical materialism with the Althusserian, realist and analytical paradigms, each of which he regards as fundamental departures from the Marxian canon." And it is here that I think it possible to conceive of historical-geographical materialism as a research programme rather than the work of any one individual.12 For what is so striking about the writings of Harvey et al. - for all their other differences - is that they have been based on a broad acceptance of the basic categories of Marx's mature political-economy: namely, the commodity, use-value, exchange-value, concrete-labour, abstract-labour, socially necessary labour time, value, money and capital and the relations between them. These Marxian categories, which capture what Postone (1996: 3) calls "the fundamental core of capitalism", remain central to their various analyses and are the basis upon which each of them has extended Marx's political-economy geographically. To be sure, there are differences of interpretation of these categories here and there, and Neil Smith in particular has developed perhaps the keenest sense of how they are necessarily complicated by the insights of feminists and other critical theorists (see, for example, Smith 1991). But I think it is the broad acceptance of these central categories that allows one to think of the work of these authors as a research programme with a structured coherence. 25 I say 'structured coherence' because these basic categories arguably form the 'core' of this research programme, or the set of basic propositions upon which it rests. But beyond this core the authors I have mentioned have each crafted distinctive and original analyses in which the various geographies of capital accumulation are elaborated in different ways and at different levels. I am not, in other words, positing any mechanical homogeneity here, and the Lakatosian notion of a research programme usefully serves to capture this sense of variety within similarity, a similarity and variety I hope to bring out in equal measure. V I . Reading theory There is no road for reading, no path or method; simply the effort and the fatigue of the difficult chance. Thomas Keenan (1993: 156) These comments about immanent critique and the importance of grasping the specificity of Marxist geographies may seem to indicate one of two things. The sceptic might suggest that, even when clearly identified, the subject of my critical inquiries is still irredeemable since it is, after all, the most 'unreconstructed' of all Marxist geographies. Alternatively, the over-enthusiastic sympathiser might suggest that once the theoretical terrain to be considered has been 'properly' delimited the 'true' nature of historical-geographical materialism can be revealed and the critics rebuffed. However, I think neither claim follows. While I certainly do not think the work I consider congenitally flawed, my intention is not to recover a 'correct' interpretation in order to counterpose it to that of the critics. I say this for two reasons. Firstly, as already indicated, I think many of the criticisms voiced against the likes of Harvey - though one-sided - have a textual grounding in theoretical claims he and his fellow-travellers have made. But more importantly, I also think such a naive supposition underestimates the importance and the difficulty of 'reading theory'. I use the textual metaphor deliberately. In one of the most celebrated discussions of the difficulty and the challenge of 'reading' any body of theoretical work, Althusser 26 (Althusser & Balibar 1970) demonstrated that reading is not only an active process - the texts never simply speak for themselves - but one of enormous consequentiality for the corpus considered. In particular, his notions of a 'symptomatic reading' and 'problematic' showed with tremendous elan that texts - in this case those ofthe late Marx - which others had interpreted in this way could be subjected to a reading which disclosed an entirely other theoretical apparatus. This insight still strikes me as exemplary - notwithstanding Althusser's declension into assertions that his reading of Marx was 'authentic' where others were demotic. It will have a particular relevance to everything that follows, because I will argue that it is possible to develop a reading of historical-geographical materialism as a problematic quite different to that identified by the critics. This is a bold claim and as likely to arouse suspicion among those already critical of Marxist geography/ies as it is possible hostility among the authors whose work I consider: for both groups might find my reading tendentious. There are no cast-iron guarantees against this. But, then again, there are also limits to what can be retrieved defensibly from the texts considered. Accordingly, to the extent that I depart from the former's critical reading and risk interpreting the latter against the grain, I try to do so responsibly and always immanently. That is, I believe the problematic I disclose lies genuinely latent within the work I consider, and I tease it out, not through any daring generalisations, but through close and careful analysis. Furthermore, to the extent that I also disclaim the notion of a 'proper' reading, I see no reason why the one I propose be regarded as necessarily ersatz, illegitimate or implausible. VII. Underlabouring ... it is ambition enough to be employed as an under-labourer in clearing the ground a little ... John Locke (1959: 14) If these comments clarify and qualify the sense in which this thesis is a 'reinterpretation' of historical-geographical materialism, they do little to address another issue which arises. 27 For this reinterpretation is evidently a dependent one. What I mean by this is that it relies on the existence of a given body of theoretical work - historical-geographical materialism -which it then reassesses. It thus makes no 'first-order' knowledge claims of its own, but rather 'second-order' claims based upon and about the epistemological practices, ontological insights and explanatory theses offered by the theoretical corpus it reconsiders. This raises several concerns, among them that my efforts are derivative, that their second-order nature debars them from making any claims about contemporary capitalism and that, in any case, their reliance on the works of others is both unjustified and arbitrary. These are all serious issues, but not intractable ones. I can approach them by refining and extending my comments on 'reading theory' to say that my reinterpretation of historical-geographical materialism can be seen as a form of 'underlabouring' on its behalf. I borrow this Lockean metaphor from philosopher of science Roy Bhaskar (1989a), who uses it in the strong sense of "clarifying and explicating what it is the sciences do and how they do it, as well as, on occasion, criticising existing scientific practices for failing to meet the standards of scientificity they set themselves" (Sprinker 1991: 123). This is 'strong' insofar as Bhaskar (op. cit. vii) is seeking a "kind of 'clearing' of the ideological ground" for the social (and natural sciences) in order to elaborate 'transcendental' claims about what the world must be like if scientific practice is to make sense. While I sympathise with the intent of Bhaskar's labours, I remain sceptical of the absolutism of transcendental argumentation, for it seemingly mimics in a different register the Althusserian distinction between an ideological and a non-ideological reading of social scientific practice, a distinction to which I just objected. However, that said, Bhaskar's Lockean self-description is in at least one respect instructive and illuminating in relation to the reading I pursue here. I mean this in the sense that Bhaskar rejects the notion of a 'first' or a priori mode of philosophical reasoning about the world. He does so because such reasoning is, on its very own terms, dogmatic and circular - Althusser's 'theory of theoretical practice' perhaps being the most notorious 28 Marxist example. For this reason, Bhaskar argues that the only sensible and defensible alternative is an a posteriori mode of inquiry in which philosophical claims about social and natural reality are pursued through an analysis of the practices of actually existing social and natural sciences. As Jeanne Schuler (1996: 184) puts it, "philosophy - like swimming - is best learned from inside the activity, not from preambles". This strategy, it seems to me, cleverly highlights the legitimacy and necessity, but also the insufficiency, of underlabouring on behalf of extant social and natural sciences. It is necessary in that social scientific practitioners - as well as their critics - may not always see clearly, or may see only one-sidedly, the epistemological nature and ontological implications of their knowledge claims. But it is at the same time insufficient in that it is no substitute for first-order theoretical and empirical inquiry.13 VIII. The plan of this thesis I had thought to give a map ... Not for me to deny the pain and pleasure of the journey to other muleteers. R. S.Neale (1985: xii) My re-reading and reinterpretation of historical-geographical materialism proceeds in four distinct but related stages. However, since the logic of this four-fold argumentative progression is by no means straightforward I need to spell out its nature and rationale. In the next chapter I round off this first Part with a more in-depth explication of the reading of historical-geographical materialism I am arguing against. I do so for several reasons: first, to present the best possible case for the critics' complaints; second, to give those complaints a more secure textual grounding than I think they presently have; and third, and most importantly, to specify precisely the claims against which my own reintepretation is ranged. However, the way I set about the task may seem unusual: for rather than examine further the arguments of particular critics I choose instead to assume the role of critic myself. Specifically, I examine David Harvey's most accomplished theoretical treatise, The Limits to Capital, with a view to teasing out and scrutinising its 29 modern - or what I call 'traditional' - architectonics. I do so because, as I noted earlier, it seems to me that few critics have grounded their objections in a truly immanent reading of their targets. Admittedly, my assumption of the role of critic may seem counter-intuitive, given my declared intentions in this work to underlabour on behalf of historical-geographical materialism. But the point is to specify as precisely as possible the problems to be addressed if a more positive reading of historical-geographical materialism is to be persuasively put forward. In a sense, then, chapter 2 serves as the straw man (sic) for the remainder of the thesis. But the analogy is imperfect, because the modern/'traditional' modality of Marxism I detect in Harvey's work is no mere caricature but, rather, an accurate reflection of particular features, or 'moments' of his work. However, that these moments do not exhaust his work - or that of the other authors I consider - is, of course, a vital pre-condition for the development of the alternative reading I propose. In Part III then move to the first stage of my reinterpretation by re-examining what I identified earlier as the 'core' of historicalTgeographical materialism as a research programme. This core is drawn directly from Marx's political-economy and is constituted by a set of well-known, key theoretical concepts. Accordingly, Part II necessarily 'returns' to Marx as part and parcel of the re-appraisal of historical-geographical materialism's core. My aims are two-fold. First, I re-examine the nature and practice of theory, in other words the epistemological status of these concepts. Second, I also re-examine the ontological claims about capitalism which follow from these concepts. In short, then, I re-appraise both the basic ontological forms of capitalism as posited by historical-geographical materialists and the epistemological/theoretical practices through which the latter grasp those forms. At least two possible objections suggest themselves here. The first is that all this talk of epistemology and ontology is overly 'philosophical' and thus tangential to the real issues. However, I hope to persuade the reader otherwise. For Part II amounts to nothing less than a reinterpretation of the fundamental features of the capitalist system in which we 30 still live today and of how we might responsibly grasp those features theoretically. The second objection is that a return to Marx is both diversionary and redundant in relation to such a reinterpretive exercise. Again, however, I beg to differ. For, as I have said, not only is Marx's basic political-economic model of capitalism still absolutely central to historical-geographical materialism; additionally, it is remarkable how few recent critics of Marxism in geography - and even historical-geographical materialists themselves - have appreciated the rich epistemological and ontological insights Marx's work possesses for the development of a distinctively non-modem, non-traditional political-economy relevant to the late twentieth century. My re-assessment of historical-geographical materialism's core begins, in chapter 3, with a consideration of both the power of and the limits to theory with what I regard as a long overdue study of epistemology and its specific relation to the protean notions of 'science' and 'dialectics'. The neglect of these topics by geographical critics - and indeed the failure of even advocates to draw out their implications - has arguably obscured vital areas of theoretical practice. Reconstructing Marx's epistemological protocols, his notion of social 'science' and his 'dialectical' method as articulated in David Harvey's work, I show that Harvey's political-economy can be defensibly read as embodying a highly sophisticated procedure with which to examine capitalism's core, one which refuses the identity of thought and the real even as it insists on the explanatory power and world-disclosive capacity of the categories it unfolds. This is not to say that Harvey presents this procedure in quite the way I present it - if he did, there would be no need for me to tease it out. Indeed, I arguably take Harvey in directions he only tentatively seeks to venture. But, the arguments I present can, I think, nonetheless be coherently drawn out of his writings. I should also say, by way of a warning, that the triple-headed discussion of epistemology, science and dialectics which comprises chapter 3 is long and demanding. However, I believe this is unavoidable if its value is to be fully appreciated. 31 In chapter 4 I move away from epistemological considerations to those of social ontology. I say 'social ontology' because my concern is not with timeless truths about the nature of reality but, rather, with the historically specific core features that constitute capitalist societies as capitalist. Through a reconsideration of both Harvey's and Neil Smith's understanding of value relations I propose a fundamental reinterpretation of these basic features. An unfavourable reading suggests that both authors instantiate capitalism as a 'closed totality' of fetishised social forms underlain by unobservable processes ultimately generated by a 'meta-subject' (the working class) located in production. Against this modern/traditional view of system and subject, I propose an alternative reading which posits capitalist social relations as distinctive, structured but fundamentally open rather than closed and not grounded in the activities of a singular meta-class agent. This reading is only seemingly paradoxical, and I pursue it through a careful reconsideration of the categories of concrete- and abstract-labour, which both critics and advocates of historical-geographical materialism have for too long either ignored or taken for granted. The view of capitalism which emerges is, I think, as eye-opening as it is undogmatic, and as arresting as it is unexpected. It may be thought that to move away from modern/traditional notions of capitalism as a self-contained, all-encompassing totality grounded in class exploitation robs Marx's political-economy - and by implication that of Harvey et al. - of its essential theoretical identity. However, what makes my argument distinctive is that I seek to retain its venerable explanatory categories but also to reconfigure their cognitive meaning. In particular, while I insist that capitalism still depends upon the exploitation of living labour, I abjure older notions of class. Rather than seeing class as a singular and essential identity, I prefer to see it as a certain positionality in relation to processes of capital circulation and accumulation. 'Glass' thus becomes radically heterogenous - an unparalleled field of difference - or what Byrne (1995: 127) usefully calls the "social proletariat" - in which diverse workers are conjoined. At the same time, while exploitation of this heterogeneity 32 provides the source of labour value, I also insist that capitalism's systemic qualities make it dominative as well as exploitative. In other words, the 'moments' of capital circulation outside the workplace are just as important as exploitation in the workplace. The upshot is that capitalism emerges as a 'system' irreducible to class and exploitative production as traditionally conceived. Instead, it can be seen as a dominative and exploitative system each of whose several moments are important and which, because it is predicated on the exploitation and domination of a plurality rather than a singularity is, inter alia, permeated by its putative 'exteriors' and by 'difference'. Even more than with chapter 3, the arguments I develop in chapter 4 can hardly be said to apply in any straightforward sense to the work of Harvey and Smith. Again, my intention is not to claim that either author secretly holds to the claims I make about capitalism. Rather, I merely aim to identify non-modern/non-traditional moments in their work and then to develop them into an explicit and worked-up account - an account neither author ventures to make more than implicit and latent. Together, then, chapters 3 and 4 deliberately reread the 'core' of historical-geographical materialism as a research programme and in so doing they reappraise the very , nature of capitalism's fundamental forms - which is why I title Part II 'Reclaiming Capitalism'. These two interventions, the one epistemological/theoretical, the other ontological, are intended to complement one another and to generate a mutually reinforcing interpretive weight. The effect, I hope, is to reveal the first level of a problematic which can make critically visible global economic relations, but in a way that their ontological forms and the theoretical means of disclosing them are anything but self-sufficient and decided.14 In Part III I then move to a second stage of my reinterpretation. Specifically, I explore how the re-reading of both the 'core' of historical-geographical materialism and of the capitalism it studies put forward in Part II can animate some of the major theoretical innovations that have marked historical-geographical materialism out as a distinctive and developing research programme. Here, then, I at last come to the 'geography' in historical-33 geographical materialism. Given the range of these theoretical innovations this exploration is necessarily partial. It revolves around three themes: the 'production of space', the 'production of nature' and, less familiarly, the 'production of subjectivity'. These themes, though they hardly exhaust the field of historical-geographical materialist inquiry, are each significant and intrinsically interrelated. In each case I try to show how my reinterpretation reconfigures Harvey et al.'s approach in a way that retains its explanatory power but circumvents 'bad' meta-narrative practice. I begin with a deliberately familiar and venerable theme - the production of space (ch. 5) - but then move on to more recent concerns - the production (and destruction) of nature (ch. 6) - and finally less familiar ones - the production of 'the subject of capital' (ch. 7). I do so in order to demonstrate the thematic reach of historical-geographical materialism but also the coherence of the reinterpretation I offer - a reach and coherence commensurate to a constellation of specific social relations which today are pervasive and global in reach. Whereas Part II is densely argued and its two chapters directly complementary, Part III is deliberately looser and more reader-friendly. Rather than constantly refer back to the earlier chapters, the three chapters that comprise Part III are each relatively free standing. This does not, of course, mean they bear no relation to what has gone before: on the contrary. But their relation to Part II involves the picking up and amplification of its arguments in the very specific contexts of the debates on space, nature and the subject respectively. For this reason the chapters of Part III are written in languages and idioms reflective of those specific debates. At times, this may make them seem to the reader to depart from my earlier concerns. But their relation to those concerns is never far away and is, I hope, readily apparent throughout. Together, Parts II and III reintepret what can be called the 'explanatory-diagnostic' moment of historical-geographical materialism: that is, its anatomisation of capitalism and its geographies. In Part IV I turn away from such explanatory-diagnostic concerns to examine the 'anticipatory-utopian' dimension of historical-geographical materialism. I 34 borrow this distinction from Seyla Benhabib (1986), who argues that any theory which claims to be 'critical' necessarily embodies both moments. Yet the authors whose work I examine have been strangely silent on this second issue, reflecting the wider reluctance within political-economic geography to reflect on the question of what Sayer (1995: 33) calls "normative standpoints". This is regrettable, as any project of envisioning capitalism must, surely, not only make capitalist relations critically visible but contest them too. I thus take this opportunity to tease out, in the final chapter, the normative dimensions of historical-geographical materialism, and try to show how my earlier reinterpretation of its explanatory-diagnostic basis can render those dimensions in a sober, relevant and realistic fashion for the 1990s. IX. Serried Pre-Cautions Thought must be divided against itself before it can come to any knowledge of itself Aldous Huxley (1929: 11) This is a summary statement of intent, but it is specific enough to alert the reader to several issues which require precautionary qualification. The first relates to presentation. A research programme begins and ends with the efforts of its practitioners. Accordingly, rather than staking out a set of 'essential' features which paradigmatically describe historical-geographical materialism and then showing how each of its advocates 'live up' to this description I prefer, instead, to let historical-geographical materialism emerge through a consideration of the work of different authors. Each chapter thus interprets the work of one or several of the authors named and, in this specific way, seeks to develop an alternative reading of historical-geographical materialist inquiry. The second issue, following on from the first, relates to thematic and authorial coverage. The reader expecting an exhaustive survey of all the themes touched upon by Harvey et al. and of more or less all of their respective works is clearly expecting too much, and I am happy to disappoint them on this score. As I have said, my focus on the 35 three themes of space, nature and the subject is a fairly synoptic one in itself, and certainly a broad enough canvas upon which to illustrate the reinterpretation I propose. The final issue relates to what can reasonably be expected of a theoretical work of this kind. I am under no illusions here. Notwithstanding my comments on the value of re-reading theory and of 'underlabouring' on behalf of it, this is clearly a limited endeavour, intellectually and practically. Intellectually in that more meso-level theoretical work and concrete empirical inquiry are also vital (if not more so) to move historical-geographical materialism forward as a research programme. And practically in that their are clearly limits to the 'effects' of work of this kind, limits I will make much of in chapter 8 and in the Coda. NOTES 'Ellen Meiskens Wood (1995: ch. 2) has offered an explanation for this apparent severing of the 'economic' and the 'political' within capitalist societies. 2I borrow the phrase 'reclaiming reality' from Roy Bhaskar's (1989) book of the same name. Christopher Norris (1996) has used the same motif to structure his most recent collection of essays. 'Including Consciousness and the Urban Experience (1985a), The Urbanisation of Capital (1985b), The Condition of Postmodernity (1989a), The Urban Experience (1989b) and, most recently, Justice, Nature and the Geography of Difference (1996). There is also the co-edited The Factory and the City (Hayter and Harvey 1993). 4 Smith (1984, 1986, 1988a, 1988b, 1989, 1990, 1991, 1993, 1996a, 1996b, 1996c, 1998), Swyngedouw (1989, 1992, 1995, 1996, 1997a, 1997b, 1997c, 1997d, 1997e), Merrifield (1993a, 1993b, 1995a, 1995b, 1995c, 1996) and Merrifield and Swyngedouw (1996). 5I discuss Deutsche's critique further in chapter 3. important here, I think, have been those recent attempts to widen and complicate what economic geographers, Marxists included, have traditionally considered to 'count' as economic research topics. This has entailed both an en-culturation and and em-bedding ofthe economy' (see, for example, Thrift and Olds 1996; Lee and Wills 1997). 'This calls to mind Bowles' and Gintis' (1985: 41) laconic observation that those "who would save Marxism from economism have mistakenly attempted to save Marxism from economics". 8From time to time in this thesis I use the collective 'we'. It is a word I use with caution and it is not invoked unthinkingly. As becomes especially apparent in chapter 4, the term is catachrestic, at once lacking a signified and yet at the same time naming a common 'constituency'. 9The irony of Duncan and Ley's critique was that almost all the Marxist geographical work they castigated for being Althusserian was not at all indebted to Althusser! Doreen Massey's early engagement with 36 structural Marxism is perhaps the one notable exception. 1 "Although, as we shall see, Harvey distinguishes historical-geographical materialism from Critical realism, I think much of his own thinking is, in effect, realist. In an indirect way, chapters 3 and 4 seek to show this. "Of course, one of the problems with the kind of'mapping' of Marxist geographies I am doing here is that Marxism is a much more diffuse presence in the work of several geographers who, strictly speaking, would not consider themselves Marxist or even post-Marxist. An excellent example is Nigel Thrift, much of whose writing is coloured by his engagement with, rather than adherence to, certain Marxian arguments. 12I borrow the Lakatosian notion of a research programme in only the loosest sense. My use of the term does not, therefore, imply any commitment on my part to Lakatos's substantive theses about the nature and progress of research communities. However, what I do find congenial about the Lakatos's terminology is the suggestion that while ostensibly different research communities do interact and, at the margins, interface, the existence of a modus tullens - a set of core beliefs - means that it is still ultimately possible to group researchers within relatively distinct research programmes. Thus, while I do concede that my decision only to include four authors within the historical-geographical materialist fold may at first sight seem arbitrary, I think it is a defensible choice since to my knowledge no other significant Marxists in the discipline today hold to the core propositions and core concepts one finds in the work of Harvey et al, the most central of which is the labour theory of value. l3This qualified defense of underlabouring usefully addresses the concerns raised above. It means that my reinterpretation of historical-geographical materialism is dependent, yes, but not derivative, since it seeks to actively disclose a different problematic to that detected by critics and advocated by practitioners. It means that my reinterpretation is second-order, yes, but that since it considers a first-order body of theory it is not debarred from making claims about contemporary capitalism based on the reinterpretation of that corpus. And finally, it means that my reinterpretation is tethered to a determinate body of existing theory, yes, but that short of rejecting out of hand all the claims that this (or any other) theoretical field makes about the world this tethering is neither unjustified nor arbitrary. 14In some senses, Part II stands in relation to the rest of this thesis as Bertell Oilman's (1971) 'Philosophical Introduction' stood in relation to the rest of his seminal book on Marx Alienation. The difference, of course, is that Oilman's understanding of the 'core' of capitalism and of Marx's political-economy is ostensibly 'modern' and 'traditional'. 37 CHAPTER TWO EXPLANATORY AND NORMATIVE PRESUPPOSITIONS: HARVEY AND THE CRISIS OF 'TRADITIONAL MARXISM' ... it is clear that traditional Marxism as a critical social theory with emancipatory intent is inadequate. David McLellan (1993: 1009) I. Introduction I have argued that while the criticisms of Marxism in geography are by no means ill-founded they nonetheless frequently under-specify their target. In this chapter I therefore want to pursue further the argument that Marxism is a 'modern meta-theory', but I want to do so specifically in relation to historical-geographical materialism. In other words, I aim to deliberately draw out the 'modem' elements of this particular modality of Marxist inquiry in some detail and thus play the role of critic of these elements and this modality. Although this may seem to contradict my declared intentions in this work, it will in fact usefully serve several purposes at once. It will further concretise the triple headed critique of'modern theory'; it will, as importantly, ground the critics' charges in a clearly specified and close reading of a specific kind of Marxist geographical work; as such, it will make a sort of 'best case' for those charges; but finally, most importantly of all, it will spell out very clearly the reading of Marxian political-economy against which my intepretation in Parts II, III and IV is ranged, and so help to identify what makes that interpretation distinctive and original. In order to move beyond the level of generalities and achieve these several aims, I intend to focus on a particular author - David Harvey - and a specific text - his The Limits to Capital. I make these choices with good reason. Harvey was, of course, the effective instigator of historical-geographical materialist inquiry and remains its most determined and important expositor. More particularly, his work has of late become emblematic of all that is worst about and all that is wrong with modern Marxian theory in human geography. The key text here, as I have already noted, is his The Condition of Postmodernity which 38 came in for considerable flak for its cognitive exorbitancy. And yet this is arguably to scrutinise historical-geographical materialism at its most idiosyncratic, not to mention weakest, point. The Condition is (with the signal exception of his most recent book) perhaps the most speculative and least rigorous of all Harvey's writings. It is not, like most of his other works, an exercise in careful political-economic theory building in the tradition of the late Marx. Rather, Harvey chooses to make grandiose and manifestly under-theorised links between the 'basal' structures of 'the economy' and the 'superstructures' of the contemporary postmodern culture he finds so objectionable. This makes the book a terrific read - indeed Harvey (unpublished) has described it as an "entertainment" - and doubtless accounts for its extremely high sales. But it also makes it a very easy target, which is why several feminist critics, in particular, have had a justifiable field-day upbraiding its meta-narrative excesses. Given this idiosyncrasy and these obvious theoretical weakenesses, The Condition is not perhaps the best place to prosecute a case against modern meta-theoretical practices. This is why I want to focus on Harvey's The Limits to Capital. Few geographical critics of Marxism in general, or Harvey in particular, have offered an analysis of this remarkable exercise in geographical political-economy. And yet The Limits is perhaps the most accomplished piece of theoretical writing in Marxist geography these last twenty years. More specifically, to repeat my earlier claim, it also stands as the charter document of historical-geographical materialism as a distinct form of Marxian inquiry. In it one finds Harvey closest to Marx, trawling his way through the Grundrisse, Theories of Surplus Value and, pre-eminently, Capital in order to restate and then extend nothing less than the whole sweep of his political-economy in a geographical direction. One commentator has rightly described it as a "complete ... exegesis, critique, and extension of Marx's mature political-economy" and it remains to this day a work of unparalleled scope and ambition within historical-geographical materialism. 39 For these reasons, this chapter focuses on The Limits with a view to reading it as a text in 'modern' Marxian meta-theory in the pejorative sense meant by the critics. As we shall see, there are considerable grounds for such a reading. But if these grounds can be pinpointed they do not, I think, amount to a fait accompli - exposing historical-geographical materialism irredeemably at one of its strongest points. Instead, more positively, I think they indicate the specific problems to be addressed and overcome if a coherent and plausible alternative problematic is to be revealed and put to work. The argument proceeds in five stages. I begin by presenting a critical interpretation of the 'mature' Marx developed by several contemporary scholars - notably Seyla Benhabib (1986) and Moishe Postone (1996) - possessed of a detailed and sophisticated knowledge of his political-economic writings. I use their work because, while these writers are extremely sympathetic to Marx's project, they nevertheless find it cognitively and normatively flawed on meta-theoretical grounds. This detour into an account of Marx is only seemingly diversionary because I then trace it through to Harvey's arguments in The Limits. Insofar as Harvey's is an avowedly 'faithful' continuation and extension of Marx, I further argue that The Limits can be plausibly seen to repeat the meta-theoretical flaws Benhabib, Postone and others expose. Seeking to construct a genealogy of these flaws, I then suggest they may be traced back to Marx's appropriation of Hegel. However, having thus presented an internal critique of the 'modern' moments of The Limits, I end on a positive, rather than a negative note, by arguing that such a critique clears the ground for a more productive reading of historical-geographical materialist work. II. The Marxian legacy? 'Traditional Marxism' Marx was, of course, a child of his time ... Roman Rosdolsky (1977: 56) Despite the forcefulness of their arguments, it is remarkable how few geographical critics of Marxism have said anything of substance about Marx himself. In mitigation, it must 40 also be admitted that these days few Marxist geographers bother, at least in print, to undertake extended considerations ofthe thinker from whom their work draws inspiration, perhaps because they now take Marx's basic claims for granted or else are more concerned with drawing out their implicit geographical dimensions. And yet, notwithstanding the wider 'crisis of Marxism', the last twenty years has seen a remarkable outpouring of new work on the nature and intent of Marx's original corpus. As I noted in the Preface, this has its 'structural' and 'analytical' components, but the work that interests me here is that undertaken by that relatively small group of political-economists outside geography who, like Harvey, regard themselves as 'classical' Marxists, or at least seek a 'classical' interpretation of Marx. For all their very real differences, these authors make a number of important basic claims about Marx's corpus. First, they suggest that although during his lifetime Marx proposed, variously, a theory of history, a political philosophy, a fragmentary theory of the state and so on, he must be seen first and foremost as a political-economist (Carchedi 1991). More precisely, they suggest that Marx must be seen as a political-economist of just one, quite specific form of society: capitalism. Second, although they reject Althusser's notion of an 'epistemological break1, many argue that Marx's most coherent and systematic political-economic work is to be found in his later texts (Moseley 1993). This is not to devalue his earlier, more philosophical writings, but it is to say that Marx's concrete theory of capitalist production is only worked out in post-1855 texts like the Grundrisse, Theories of Surplus Value and, of course, the three volumes of Capital. Third, these authors take seriously Marx's claim to be a social scientist, and argue that the categories of his political-economy are unfolded as an extremely careful attempt to grasp the specificity of capitalist social forms (Murray 1988). Finally, many of these authors agree that in his 'scientific' endeavours, Marx was not an 'economist' in the conventional sense we know it today, but a theorist of social relations and social forms. To cite Carchedi (1991: ix), "economics is first and foremost a social science, a science which studies historically specific social 41 phenomena", and an even casual acquaintance with Marx's later writings quickly indicates that he was not concerned with algebra or economic modelling. As Marx (1973: 106) himself said, the basic categories of his critique of political-economy "express the forms of being, the determinations of existence ... of this specific [capitalist] society". This is vitally important. While it does not sidestep the need for quantitative political-economy (indeed, debates over the 'transformation problem', the operationalisation of value-magnitudes etc. still proceed apace), it does mean that Marx's work should be seen primarily as what Postone (1996: 18) calls "a critical ethnography of capitalist society"! Some of these points will be familiar and uncontentious, others less so. However, they usefully contextualise a reading of Marx's political-economy as what I will call 'traditional Marxism', that is, as a 'modem' discourse in something like the sense of that term proposed by geographical critics.1 It is a reading I can pursue through the critical analyses of Postone (1996) and Benhabib (1986) - two authors who thematise Marx's work in the four ways described - and through a consideration of the pre-eminent Marxian text, Capital, especially volume 1. This claim to pre-eminence is, I think, quite justified because in it Marx's lays out the most fundamental categories of his political-economic critique -and it is for this reason that subsequent Marxists have devoted so much attention to it. Volume 1 is also distinctive in that, unlike the Grundrisse, for example, Marx's account is structured in a tightly logical and exacting manner. And it is the place where, more than any of his other published writings, Marx makes great play about seeking the "luminous summits of science". Traditional Marxism: the explanatory-diagnostic moment In his magesterial Time, Labor and Social Domination, Moishe Postone adumbrates an interpretation of Capital as a form of what he terms 'traditional Marxism'.2 It is a reading of Marx's core political-economic arguments which, he argues, is widespread among classical Marxists and to be found in texts as diverse as Lukacs' History and Class Consciousness 42 (1971), Manuel's (1978) Late Capitalism and Sweezy's (1969) The Theory of Capitalist Development. It is, furthermore, a reading which enjoys enormous textual warrant in Marx's political-economic writings. And it is is a reading now so familiar that it has almost come to stand for what the late Marx 'really' said. It goes something like this. Marx begins Capital with the category of the commodity, the simplest and most pervasive phenomenal form in which social wealth appears within capitalist societies. He then shows that commodities have both a use-value and an exchange-value and, corresponding to this, that the labour required to produce commodities must also have a two-fold aspect. However, this still raises the question of how commodity exchange can take place. Through an analysis of the 'elementary', 'expanded' and 'general' forms of exchange, Marx shows that commodities possess a dimension common to but distinct from their use-value form: that is, they possess value. Because of this, value is the general form of social wealth in commodity producing societies. However, this value dimension only remains latent unless it can be actualised by a commodity which can serve as its general measure. Hence, Marx derives the 'necessity' for money, which serves as a universal measure against which commodities can be valued and exchanged. Effectively, then, money facilitates relations between things - commodities - which are in fact relations between people, since commodities are embodiments of 'abstract labour', which is the 'measure' of value. So it is that at the end of chapter 1 of Capital Marx presents his famous discussion of the 'fetishism of commodities', where he shows that social relations take on the dissembling empirical appearance of thingly relations. However, Marx's categorial account has much further to go than this because the hidden abode of relations are not merely exchange relations between people. Instead, certain agents - namely, capitalists - enter into exchange as a means of receiving at the end of the transaction the same commodity they put forward at the start: money. The only conceivable reason for doing so is to receive more of this commodity than was put forward at the beginning of the transaction. Since this commodity is the general form of value this 43 is the same as saying that more value is sought after. In its several forms, money thus becomes much more than the medium of commodity circulation, C-M-C. Marx's 'general formula for capital' specifies a form of circulation M-C-M' where an extra increment of money - or profit - is accrued by the seller. Money which circulates in this way is what Marx calls capital. And yet this form of circulation seems impossible because the exchange of commodities assumes an equality. The only way an inequality could be derived from such a transaction is through robbery, deception, buying cheap and selling dear and the like, since no party would willingly exchange for less than they have put forward. So where, then, does profit come from? To answer this key question Marx shifts his attention from the realm of exchange to that of production. In capitalist societies, he argues, we know that there are at least two major classes involved in commodity production: capitalists and workers. Capitalist relations of production entail the buying of labour-power by the capitalist as part of the production process, along with raw materials, machinery and the like. And it is here, in chapter 6 of Capital, that Marx specifies the origin of profit: living labour. At first sight this seems contradictory. After all, since labourers sell their labour-power for a wage and since the exchange bargain is based on an equality then this surely cannot be true. However, Marx argues that labour-power is unique among commodities in that its use-value aspect enables it to create more value than it consumes, that is, surplus value. This is a complex claim which Marx seeks to make good on in several ways. Most important is his argument that when set in motion it is labour-power which is productive of both the concrete- and abstract-labour attached to commodities. What this in fact means is that labour-power is the source of value, since value is nothing but materialised abstract-labour. But this claim in turn requires explanation, since, apparently, abstract labour has no measure to actualise it. Until, that is, Marx introduces the notion of 'socially necessary labour time'. Clock time, an historical invention closely associated with capitalism's rise, is the real measure which renders concrete labours effectively commensurable and which 44 brings into being the dimension of 'abstract labour'. Marx's notion of 'social necessity' is vital here because it indicates the tendential formation of a socially average production period against which dispersed concrete labour-processes can actually be compared and brought into relation. This has several important implications. It means that capitalists have an incentive to employ labourers for a period exceeding that socially necessary for the latter to reproduce themselves. It also means that in socially necessary labour time capitalists face a real social average against which their own productive activities must compare if a profit is to be made. Finally, and most importantly, both points mean that while capitalists only pay labourers an amount necessary for their social reproduction, they employ them for an additional period effectively given gratis in which surplus value is created. Moreover, because the the wage payment is seemingly fair and occurs in exchange, this extraction of surplus labour is achieved only in production. So it is that labour-power, because of its unique commodity status, is for Marx exploited. And so it is too that Marx introduces the concepts of 'absolute' and 'relative surplus value', and devotes considerable historiographic attention in volume 1 to the intense struggle over the length of the working day. These core categorial claims - otherwise known as Marx's 'labour theory of value' -have long been the subject of tremendous interpretive controversy, both within Marxism and without (see, for example Fine [1986] and Steedman and Sweezy [1981]). But as Postone notes, Marx's political-economy does not, of course, end with these fundamental claims. On the contrary, for all its importance volume 1 cannot be considered without volumes 2 and 3 of Capital, where Marx builds towards a theory of economic crisis, focussing particularly on the tendency of the rate of profit to fall and on the related contradiction between the forces and relations of production. The 'law of value', as Marx calls it, is an historical - not general - one relating specifically to the capitalist system. What is remarkable about it is that it is generated by capitalism itself, through its internal contradictions and disequilibriating tendencies. The result is a series of acute economic 45 crises, which generate not only capital devaluation but also job loss and social misery on a very large scale. In in this sense, as is well-known, Marx believed that capitalism, paradoxically, had the capacity to sow the seeds of its very own dissolution and thus create the conditions for genuine working class emancipation. And in this sense too Capital can be seen as elaborating a 'crisis theory' whose purpose is to disclose the historical limits to capitalist social organisation. This unrefined summary of Marx's basic arguments will be very familiar to most readers. In what sense, then, is it a 'traditional' reading of Marx? Postone (1996: 8) offers an answer: "Within this general framework ... Marx's critical analysis of capitalism is primarily a critique of exploitation from the standpoint of labour: it demystifies labour in capitalist society ... by revealing labour to be the true source of social wealth, and ... by demonstrating that that society rests upon a system of exploitation". According to Postone, then, 'traditional Marxism' is characterised by four specific features. First, it is a critique on behalf of a singular subject - the working class. Second, it is a demystifying critique of capitalism which shows how the 'truth' of social relations are hidden behind 'surface appearances'. Third, it therefore prioritises the sphere of production: to the extent that surface appearances are delusive they are ultimately subordinate - in both explanatory and ontological terms - to the exploitative production 'essence' behind them. Finally, insofar as traditional Marxism is able to make these plenary claims it forgoes epistemological modesty in order to posit its theoretical representations as real-istic representations, which accurately and exhaustively capture their socio-ontological objects. Traditional Marxism: 'critique' and