David Harvey at 90, and the theory that capital produces space
For half a century the Marxist geographer has argued one thing: capital does not simply move through space, it builds space, and it escapes its own crises by rearranging what it has built. From Haussmann's Paris to the data centres of rural Louisiana, an assessment of the idea and of what it cannot do.
In June 1853 Napoleon III made Georges-Eugène Haussmann prefect of the Seine and instructed him to give Paris air, unity and beauty. Over the following seventeen years the works cleared 19,730 buildings and 120,000 lodgings and put up 34,000 structures containing 215,300 units. Rents across the Second Empire roughly trebled; the poor of the centre moved east. The first phase alone cost 278 million francs. The Péreire brothers' Crédit Mobilier advanced 24 million against the development rights on the new frontages, and by 1867 Haussmann had contrived to raise some 463 million more through bond schemes the legislature refused to tolerate. He was dismissed on 5 January 1870. Sixteen months later the Commune was shot to pieces in the boulevards he had widened.
Most accounts of Haussmann's Paris are architectural, or they are about Napoleon III. David Harvey's, in Paris, Capital of Modernity (Routledge, 2003), is about money. He reads the rebuilding as the answer to an economic problem: after the crash and the revolution of 1848, France had idle capital and idle labour and no profitable way of joining them. The Empire found one. It pushed the surplus into stone, iron, credit and street frontage, bought the better part of two decades, and generated in the process a debt load and a displaced working class that fed directly into 1871.
This is the argument Harvey has been making since the early 1970s, and it fits in a sentence. Capital does not merely operate in space; it produces space — cities, ports, motorways, pipelines, rail, housing stock — and when profitable outlets close elsewhere it rescues itself by building. The building absorbs the surplus, defers the crisis and moves it. Harvey calls this the spatial fix. He turns 91 on 31 October, and it is arguably the most portable idea Anglophone Marxism produced after 1945.
Harvey was born on 31 October 1935 in Gillingham, Kent. He read geography at St John's College, Cambridge, took a doctorate there in 1961 on rural change in Kent between 1800 and 1900, spent a post-doctoral year at Uppsala and lectured at Bristol from 1961 to 1969. In 1969 Edward Arnold published Explanation in Geography: 521 pages of rigorous, entirely respectable positivist methodology, an attempt to give the discipline the philosophy of science it was thought to lack. It made his reputation.
Four years later he published Social Justice and the City (Johns Hopkins University Press, 1973), the second half of which takes most of it apart. Academics revise; very few disown a major book at forty-two, having only just been rewarded for writing it. The occasion was Baltimore, where Harvey had moved in 1969 and where the consequences of the 1968 riots were still visible from the windows. The instrument was a 1972 essay in , 'Revolutionary and counter-revolutionary theory in geography and the problem of ghetto formation' (4(2): 1-13), which put the charge to his own side: geographers could describe the segregation of a city street by street and had nothing whatever to say about why it was there. The essay ends flatly. 'It is the emerging objective social conditions and our patent inability to cope with them which essentially explains the necessity for a revolution in geographic thought.'
Alasdair MacIntyre died in May 2025 insisting that modern moral argument is interminable because the words outlived the world that made sense of them. His diagnosis in After Virtue, the disquieting suggestion, emotivism, Nietzsche or Aristotle, the wait for another Benedict, reads differently in a year of shouting. Here is what holds and what does not.
Charles Taylor, Alasdair MacIntyre, Friedrich Nietzsche
What followed was not a retreat from geography into politics but the reverse. Marx had written a theory of accumulation in time; Harvey gave it a geography. He stayed at Hopkins until 1987, held the Halford Mackinder chair at Oxford from 1987 to 1993, returned to Baltimore for eight years, and since 2001 has been Distinguished Professor of Anthropology and of Earth and Environmental Sciences at the CUNY Graduate Center.
The Limits to Capital is the book that matters and the one nobody finishes#
The Limits to Capital (1982) is Harvey's most serious theoretical work and, by some distance, his least read. The term 'spatial fix' had appeared a year earlier in an Antipode essay pairing Hegel, von Thünen and Marx; the book gives it a mechanism. Capital circulates through a primary circuit of production and consumption and, when that clogs with unsellable goods and unemployable money, switches into a secondary circuit: fixed capital and the built environment. Roads, factories, offices, flats. Long-lived, immobile, and capable of soaking up surplus for decades.
Everything turns on the word 'immobile'. To fix capital in place is to bet that the place will still be profitable when the asset matures, and the bet frequently fails: Detroit is the textbook case, as is the British industrial north, as in a slower way is any port the container ship made redundant. So the spatial fix does two contradictory things at once, and Harvey knows it. It stabilises accumulation now and lays out the geography of the next devaluation.
Baltimore taught him the small version; China taught the field the large one. Between 2011 and 2013 China poured roughly 6.4 gigatonnes of cement, against about 4.4 gigatonnes used by the United States across the whole of the twentieth century. Read as a development story that is a statistic about growth. Read through The Limits to Capital it is what absorbing a global surplus looks like at speed.
The Condition of Postmodernity (Blackwell, 1989) carried Harvey beyond geography. Its argument is that the cultural shift catalogued by the postmodernists — fragmentation, pastiche, the collapse of grand narratives, the thinning of historical time — was the phenomenology of a change in the circulation of capital. Under the flexible accumulation that followed the crisis of 1973, turnover time fell, distance stopped mattering in the old way, and the experience of space and time was compressed accordingly. He called it time-space compression. The book has accumulated something near 50,000 citations and was named by The Independent among the fifty most important works of non-fiction published since 1945.
It also produced two of the sharpest reviews of the period, in the same issue of Environment and Planning D: Society and Space in March 1991. Doreen Massey's was called 'Flexible Sexism' (9(1): 31-57). Taking Harvey's book together with Edward Soja's Postmodern Geographies, she argued that both authors had mistaken a particular standpoint for a universal one, had handled feminist scholarship as an afterthought, and had relegated the concerns of ethnic minorities and of the non-Western world, as she put it, to parentheses. Rosalyn Deutsche's 'Boys Town' (9(1): 5-30) went at the epistemology: Harvey's claim to see the objective totality beneath the fragments entails a refusal of feminist theories of representation, and treating difference as the ideological surface of late capitalism conveniently disqualifies the people raising it. Meaghan Morris followed in Theory, Culture and Society in 1992 with 'The Man in the Mirror'.
Harvey has never really conceded the point, and the cost of not conceding it recurs. Reviewing Seventeen Contradictions and the End of Capitalism (Profile, 2014) for Society and Space, Ståle Holgersen asked why nature is internal to capital in Harvey's scheme while gender and race are external to it. Twenty-three years after Massey, that is the same objection in a different coat.
A Brief History of Neoliberalism (Oxford University Press, 2005) is Harvey's most widely read book and his most contested. Its thesis is compressed into a page. Neoliberalisation, he writes, can be read 'either as a utopian project to realize a theoretical design for the reorganization of international capitalism or as a political project to re-establish the conditions for capital accumulation and to restore the power of economic elites'. He then says which: 'In what follows I shall argue that the second of these objectives has in practice dominated.' The free-market theory, on this reading, was mostly justification.
The evidence he assembles for the distributional claim is substantial and the claim has aged well. The causal claim is weaker, and critics on his own side have said so: reading a coordinated class project off an outcome is a functionalist move, and Harvey's account of who decided what, when, and in which room is thin where it needs to be thick. Attributing to intention what may be the sum of many separate defensive manoeuvres is a standing hazard of the method.
Marx's primitive accumulation is a story about origins: the enclosures, the clearance of the commons, colonial plunder, the violence that produced a propertyless class before capitalism proper could begin. In The New Imperialism (Oxford University Press, 2003) and a companion essay in the Socialist Register, Harvey argued that this is no phase but a permanent department of the system, and named it accumulation by dispossession.
The examples are unglamorous and everywhere: the privatisation of water and rail, the enclosure of pension entitlements, the patenting of seed stock and of genetic material, the conversion of housing from a thing people live in to an asset class people store value in, the sale of public land at valuations assuming it will never be needed again. The mechanism here is not the extraction of surplus value from wage labour but the transfer of existing assets into private hands at less than they are worth, and Harvey's proposal is that when the primary circuit stalls this becomes the more attractive route. Hence the decades after 1980.
A category this wide, though, stops cutting. If enclosure, austerity, gentrification, student debt and data scraping are all dispossession, the word is doing classificatory rather than explanatory work, and the differences between those things may matter more than what they share.
Harvey began teaching Capital to graduate students at Johns Hopkins in the early 1970s and has taught it ever since. In 2007 a CUNY graduate student, Chris Caruso, offered to film the course; Harvey has written that he agreed without expecting anyone to care. The videos went up free in 2008, the year the financial system came apart, and found an audience no university had assembled for a close reading of a nineteenth-century economics text: activists, architects, planners, trade unionists, undergraduates in countries with no Marxist tradition, and a great many people trying to work out what had happened to their mortgage. Two print companions followed, in 2010 and 2013, then volumes on the Grundrisse and on Capital's later books.
Publishers and news reports call the audience enormous and the series a MOOC before the term existed. No authoritative cumulative figure for views or downloads appears to exist, on Harvey's own sites or elsewhere, so the honest statement is that the lectures have been widely watched for eighteen years and the number is not on the record. What can be said flatly is that Harvey is one of very few academics of his generation whose primary readership is not academic, and that he chose this at seventy-two. The Story of Capital: What Everyone Should Know About How Capital Works (Verso, 24 February 2026, 400 pages) is the distillation.
The most substantial internal criticism concerns the spatial fix itself. Bob Jessop, in David Harvey: A Critical Reader (Blackwell, 2006, pp. 142-166), identified three problems: Harvey uses 'fix' both literally, for capital immobilised in physical form, and metaphorically, for a temporary patch on a crisis tendency, without settling whether these are one phenomenon or two; he treats spatial and temporal fixes additively rather than as aspects of a single process; and he builds the apparatus on one contradiction, capital's need to be fixed and liquid at once, while other contradictions go untheorised. An Antipode special issue marking the book's twentieth anniversary, introduced by Neil Brenner, Noel Castree and Jürgen Essletzbichler (36(3): 401-406, 2004), made related complaints; Erica Schoenberger's contribution revisited the concept directly.
Paul Mattick's review of the 2006 reissue, in Historical Materialism 16 (2008): 205-232, goes after the crisis theory. Harvey, he argues, confuses levels of abstraction, treats crisis as a failure of equilibrium when Marx's point was that capitalism has no equilibrium to fail at, and reaches for state credit and surplus absorption in a way that is Keynesian rather than Marxist. One need not accept Mattick's own position to grant the diagnosis: Harvey is under-specified about when a switching crisis occurs, how severe it will be, and what would count as evidence against the model.
Then there is the record. Harvey has been analysing the terminal contradictions of capitalism since 1973 and capitalism is still here. Two things should be said in fairness. He does not, in fact, predict collapse: he describes his method in Seventeen Contradictions as using the power of abstraction to build a model, and asks whether certain contradictions 'may prove fatal' — the modal verb is his. And a theory whose central claim is that capital survives by displacement was never going to generate collapse predictions; it generates predictions about where the next devaluation lands. The harder version of the objection survives both defences. A framework that can accommodate any outcome after the event is not being tested by events, and the spatial fix has a way of explaining everything and forecasting nothing in particular.
The phrase most often attached to Harvey's name is not his. Le Droit à la ville was published by Henri Lefebvre in 1968, and Harvey has been clear about the debt since 1973, when the final part of Social Justice and the City closed on Lefebvre's argument. What he added, in a 2008 essay in New Left Review and then in Rebel Cities (Verso, 2012), was the link to surplus absorption: if urbanisation is how capital disposes of what it cannot otherwise sell, then control of urbanisation is control of the social surplus, which makes it a question about democracy rather than about planning. His formulation: 'The right to the city is far more than the individual liberty to access urban resources: it is a right to change ourselves by changing the city.'
What follows is an extension of Harvey's framework rather than a claim he has made; I can find no sustained treatment of artificial intelligence infrastructure in his published work.
The five largest American cloud and AI providers — Amazon, Alphabet, Microsoft, Meta and Oracle — have guided to between 660 and 690 billion dollars of capital expenditure in 2026, against roughly 380 billion in 2025. The United States has upwards of 4,000 data centres operating or under development. They consumed 183 terawatt hours of electricity in 2024, more than four per cent of national demand, and are projected to reach 426 by 2030. Virginia alone has 643 of them, taking about 26 per cent of the state's electricity in 2023.
Individual projects make the case better than aggregates. Meta's Hyperion campus in Richland Parish, Louisiana, covers close to 4,000 acres, roughly eleven square kilometres, or four times Central Park. The announced figure is 200 billion dollars. It needs about five gigawatts of computing capacity, ten new gas-fired plants and 240 miles of transmission line, and is due in 2030. Four-fifths of it belongs not to Meta but to Blue Owl Capital, through a special-purpose vehicle named, with some local feeling, Beignet; BlackRock has bought over three billion dollars of the bonds. When complete it is expected to support more than 500 permanent jobs.
Every element of that is the object Harvey's theory was built to handle: a very large quantity of surplus capital, immobilised in fixed assets with a thirty-year life, in a specific and not especially wealthy place, financed by claims on future revenue rather than present output, generating almost no employment relative to outlay, and reorganising a regional power grid around itself. A spatial fix with fibre optics. Whether it is also the setup for a devaluation depends on whether the demand appears, and nobody, Harvey included, knows that. The framework is not prophecy; it tells you which questions are worth asking, and they are questions about land, water, electricity, debt maturities and who holds the asset when the lease expires. Our coverage of artificial intelligence and the shape of capitalism and of Marx on alienation and the commodity comes at the same terrain from other directions.
Harvey's achievement was to make geography a condition of critical theory rather than a decoration on it, and to insist that a map records decisions. Read alongside Immanuel Wallerstein's world-systems analysis, the two are the most durable thing post-war Marxism did with the problem of scale: Wallerstein took the world as his unit, Harvey took the street, and neither would accept the nation-state as the natural container of economic life. The theory has holes and its author has not filled several of them. It remains the best available account of why capital keeps having to build.
Yanis Varoufakis says capitalism is dead and that 'cloud capital' now extracts rent from digital fiefs rather than profit from markets. We set out the technofeudalism thesis as an argument to be tested, then weigh the strongest objections from Morozov, Snow and Gane, and Varoufakis's 2026 reply to his Marxist critics.
The Chinese-American philosopher Tu Weiming has spent sixty years pressing a single claim: that to become modern a society need not become Western. This profile sets out his idea of 'cultural China', the Boston Confucians, and his attempt to revive Confucian humanism as a living philosophy, and asks whether it holds.
.jpg?width=500 "Kant")
Comments
0Loading comments…