James O'Connor's second contradiction of capitalism

The Deepwater Horizon disaster: capital undermines its ecological basis

A critical introduction to the influential ecological Marxist concept of 'the second contradiction', proposed by James O'Connor.

O'Connor's idea of 'the second contradiction of capital' is one of the most influential in ecological Marxism, and remains an important point of reference. In 1988 O'Connor was a founding editor of the quarterly journal Capital, Nature Socialism, and in an article in the first issue he introduced the concept of the second contradiction.

O'Connor's argument falls on the cusp between Marxist orthodoxy on the one hand, and both growing ecological awareness and the emergence of 'new social movements' on the other. The concept of the second contradiction is developed therefore both to preserve, and to develop traditional Marxism into new domains (ecology) and for new subjects (such as feminist, environmental, and national movements). But to do so, he has to relax some of the more orthodox received wisdom. He draws inspiration from Karl Polanyi's argument, in The Great Transformation, that market relations come into conflict with social reproduction. Combined with Marxism, this provides an alternative to:

...bourgeois naturalism, neo-Malthusianism, Club of Rome technocratism, romantic deep ecologism, and United Nations one-worldism.

We'll now sketch out the what the 'first' and 'second' contradictions actually are, before engaging with two of O'Connor's key ideas: an 'underproductionist' theory of crisis, and 'the rebellion of nature' thesis.

A tale of two contradictions1

It's been said that the 21st century is a race between the first and second contradictions. 2 The first, central to orthodox Marxism, is the contradiction between the forces of production on the one hand, and the relations of production on the other. The forces of production include things like forms of labour, scientific knowledge, co-operation, and technology, while the relations of production are the class relations arising out of private property in the means of production.

In orthodox Marxist theory, this contradiction, via the labour movement, was held to drive a more or less linear progression towards more centralised ('socialised') industry and more planning, by states or industrial-financial monopoly capitalists (restructuring "into more transparently social, and hence potentially socialist, forms"). Socialism could then be understood as the workers, or in practice, the Communist Party, taking over the apparatus of centralised production and planning that capitalism had bequeathed.

O'Connor flirts with this orthodoxy but isn't satisfied with a unilinear, teleological history. Rather, he insists that:

it has become obvious that much capitalist technology, forms of work, and the like, including the ideology of material progress, have become part of the problem not the solution.

He subsequently identifies a second source of capitalist crisis, a second contradiction. This is between the forces and relations of production combined and the conditions of production, broadly speaking, between political economy and the environment:

An ecological Marxist account of capitalism as a crisis-ridden system focuses on the way that the combined power of capitalist production relations and productive forces self-destruct by impairing or destroying rather than reproducing their own conditions.

A crisis of underproduction?

To develop this claim, O'Connor seeks to develop a theory of crisis adequate to the second contradiction in the way Marxists had for the first. It's probably helpful to clarify some Marxist jargon here, so that we can see where O'Connor's originality lies. We’ve tried to keep it simple in the text and pushed some detail to the footnotes.

Perhaps the most common crisis theory we hear is underconsumptionist. Strictly speaking this is more of a left-Keynesian theory than a Marxist one, but many Marxists espouse something along these lines (such as Rosa Luxemburg). The central claim of an underconsumptionist crisis theory is that it's profitable for individual capitalists to drive down wages, but yet when all capitalists do this, it suppresses consumer demand, since poorer workers are less voracious consumers. Thus capitalists are caught in a rationality trap. The solution is usually some form of social partnership, via the state, trade unions, or both, to keep short-sighted capitalists from undermining aggregate demand, and thus their own interests. Ha-Joon Chang is one of the most prominent advocates of this position.

Therefore for underconsumptionists, crisis is not inherent to capitalism, it can be avoided by more enlightened self-interest and social partnership, making capitalists realise that 'decent wages' are win-win.3 The more properly Marxist theory, by contrast, is one of overproduction. This view points out it's rarely in the interests of capitalists to pay higher wages, since it eats into their profits. And it also points out an error in underconsumptionist thinking: consumer demand isn't the only demand, firms also consume, e.g. purchasing new machinery. Of course, firms only invest in the hope of future profits, so this creates a 'grow-or-die' dynamic, but it means that underconsumption is a symptom rather than a cause of crisis.4

Rather, for the overproductionist point of view, firms investing in more mechanised processes each seek advantage over their rivals by becoming more productive. But as these productivity gains generalise across the industry, rates of profit begin to fall. For overproductionists, this is the rationality trap: capitalists are compelled to become more productive to compete, but increased productivity and competition ultimately depress profit rates. Eventually, suitably profitable investments begin to dry up. Too much capital chases too few profits. Firms either engage in more speculative activity or withhold investment, the economy falters, some firms fail and others are devalued. The general devaluation of capital – crisis – helps restore the rate of profit, and then a new round of accumulation can begin.5

It is in this context that O'Connor proposes a different kind of crisis: underproduction. 6 His argument is as follows: capital increasingly dominates and incorporates its environment. This includes the natural environment – with things like soil nutrient depletion and deforestation. This requires things like synthetic fertilisers or planned forestry. It also includes the social environment – peoples' health suffers from pollution and overwork, which necessitates things like healthcare provision to maintain the workforce. In this way, nature itself becomes produced, so "that 'natural barriers' may be capitalistically produced barriers, that is, a 'second' capitalised nature." 7

Consequently, capital increasingly has to produce its own environment, and this represents an overhead cost at the system level. A crisis of underproduction thus includes, but exceeds, a crisis of social reproduction, encompassing both social and environmental reproduction.

...we can introduce the possibility of capital underproduction once we add up the rising costs of reproducing the conditions of production. Examples include the health care costs necessitated by capitalist work and family relations; the drug and drug rehabilitation costs; the vast sums expended as a result of the deterioration of the social environment (e.g. police and divorce costs); the enormous revenues expended to prevent further environmental destruction and to cleanup or repair the legacy of ecological destruction from the past... 8

In light of the massive costs associated with mitigating and adapting to climate change, this line of argument seems highly salient. This also provides a framework to understand the contemporary discourse around 'ecosystem services': once natural cycles are perturbed and/or replaced by political-economic ones, the price of nature's 'free gifts' is revealed in rising financial costs. Examples could include the use of (energy intensive) synthetic fertilisers to compensate for declining soil fertility, or the need to hand-pollinate crops where intensive pesticide use and loss of hedgerow habitats have wiped out pollinators. For O'Connor, this dynamic feeds back into economic crisis:

No one has estimated the total revenues required to compensate for impaired or lost production conditions and/or to restore these conditions and develop substitutes (...) all unproductive expenses from the standpoint of self-expanding capital. 9

The rebellion of nature

O'Connor therefore explains the rise of new social movements as a response to the crises of underproduction. This seems to stretch the concept too far, as well as being politically problematic in framing feminist, indigenous, or anti-racist movements as 'nature' rebelling. However, the narrower claim that capital's encroachments on the environment solicit resistance is surely borne out by today's anti-pollution riots, fracking resistance, and the No TAV movement.

The combination of crisis-stricken capitals externalising more costs, the reckless use of technology and nature for value realisation in the sphere of circulation, and the like, must sooner or later lead to a 'rebellion of nature', that is, to powerful social movements demanding an end to ecological exploitation.

It is perhaps here that O'Connor's attempts to preserve Marxist orthodoxy through broadening its scope comes up against its limits. It is true that ecological explanation can and does provoke diverse ecological struggles. But this is no more inevitable than austerity necessarily giving rise to a powerful, unified class movement. It provides the grievance, but grievance is only one of the conditions for a movement. In the case of climate change, we also encounter a further problem. With a polluting factory or mine, the effects are often felt most harshly in the immediate vicinity, or at least in specific localities, e.g. down river, and the origin is relatively easily identified. But climate change is massively nonlocal in both space and time. 10

That is to say, as the first official climate change refugees evacuate their drowning island, who is to blame? Aside from the impossibility of attributing it to any specific greenhouse gas emissions, the inertia of the climate system means the emissions that caused this warming happened decades ago (warming from recent emissions is still 'in the pipeline'). How can a movement coalesce against such a diffuse threat, one that is so massively distributed in space and time? We are not saying this is impossible, only that it can't be taken for granted that 'sooner or later' powerful social movements will cohere.

In a sense, this is no different to the more narrowly defined class struggle. Capitalism generates grievances daily, a small percentage of these flare up into local struggles, a small percentage of these broaden out or link up into wider movements, and a small percentage of these escalate to shake the whole social order. Fragmentation and multiplicity are the norm. Ecological struggles seem to follow the same pattern. Indeed, O'Connor insists that "issues pertaining to production conditions are class issues (even though they are also more than class issues)." Climate change will certainly escalate the grievances, with crop failures, rising food prices, and displaced populations to name but three.

Reading O'Connor in the context of climate change remains productive, even if it's sometimes a case of working out where he goes wrong. Indeed, when he writes that "atmospheric warming, acid rain, and pollution of the seas will make highly social forms of reconstruction of material and social life absolutely indispensible", we can only agree.

  • 1. If counting contradictions floats your boat, David Harvey's new book is called 17 contradictions of capitalism. The editorial of Endnotes 3 goes the other way, arguing that there's only one contradiction, between use value and exchange value, which however gives rise to numerous antagonisms.
  • 2. Though we can't find the source of this quote: if you know please tell us!
  • 3. This is a simplification, for example Rosa Luxemburg’s underconsumptionist crisis theory argued crisis was inherent to capitalism, but could be deferred by expansion into non-capitalist areas – imperialism. But once the whole world was carved up by imperialist powers, the crisis would become inescapable.
  • 4. For reasons of brevity, we are simplifying a large and contentious body of crisis theory here. For example, underconsumptionists counter that all demand is, ultimately, consumer demand.
  • 5. Nothing in this account requires a terminal crisis, theories of which have more to do with political requirements: either to rationalise the ‘evolutionary socialism’ of social democracy, or to compensate for a proletariat not living up to its ascribed revolutionary role.
  • 6. Apologies for the succession of arcane jargon - unfortunately it goes with the territory. O’Connor is developing an observation made by Marx here: “The greater the development of capitalist production, and, consequently, the greater the means of suddenly and permanently increasing that portion of constant capital consisting of machinery, etc., and the more rapid the accumulation (particularly in times of prosperity), so much greater the relative over-production of machinery and other fixed capital, so much more frequent the relative under-production of vegetable and animal raw materials.”
  • 7. This argument parallels the move from formal to real subsumption of the labour process, which is Marx's account of how capital remakes production in its image - deskilling, organising production lines and global value chains etc.
  • 8. O'Connor follows Marx here: "The maintenance and reproduction of the working class is, and must ever be, a necessary condition to the reproduction of capital. But the capitalist may safely leave its fulfilment to the labourer’s instincts of self-preservation and of propagation." Clearly treating divorce as 'social deterioration', and implicitly seeing social reproduction as the realm of natural instincts elides a host of gendered dynamics.
  • 9. Subsequently there have been attempts to forecast the economic cost of climate change, for instance the Stern Review, and the reports of the IPCC’s Working Group II.
  • 10. This phrasing is borrowed from Timothy Morton's Hyperobjects: philosophy after the end of the world.

Posted By

Out of the Woods
Apr 25 2014 08:32


  • Capital increasingly has to produce its own environment, and this represents an overhead cost at the system level. In light of the massive costs associated with mitigating and adapting to climate change, this line of argument seems highly salient.

Attached files


Jun 19 2017 18:08

I think that, far more than most, this is a fair reading/presentation of O'Connor's approach. What follows is probably as much a slightly different account of Jim's theory as a personal variant - I was his student as CNS got off the ground and among the few to use the second contradiction as an organizing heuristic for my dissertation.
When you write: "O'Connor therefore explains the rise of new social movements as a response to the crises of underproduction. This seems to stretch the concept too far, as well as being politically problematic in framing feminist, indigenous, or anti-racist movements as 'nature' rebelling." it seems to me that you elide the difference between "nature" and "conditions of production." While it is true that capital naturalized the necessary conditions of production it, or the state and civil society, produces it is not true that conditions of production = nature. O'Connor's argument is much more in line with socialist feminist accounts of the ways in which capitalism reworks, and often exacerbates, gender divisions of labor in social reproduction (see how Nancy Fraser has recently picked up the second contradiction along these lines).
Along these lines, whereas the New Left sought to democratize the state bureaucracies charged with reproducing the ecological, personal and communal conditions necessary for efficient commodity production and realization - a moment in the state's legitimation "function" - new social movements sought to democratize the sciences, families and communities associated with the reproduction of those same conditions. Martin O'Connor, encouraged by the ecofeminist Ariel Salleh, argued that the flip side of capital's conditions of production was new social movements' conditions of life. Neither are natural, yet capital and the state naturalize both conditions of production and conditions of life.
Again along these lines, in terms of the material in footnote 8, divorce isn't necessarily social deterioration but it is a deterioration in - a result of contradictions embedded in - the conditions of personal reproduction produced and legitimated by capital and the state... particularly within the post-war professional suburban landscape. All of the elements of the second contradiction are to describe contradictions in the conditions of production/life capitalism and the state produce for capital. It is only if the ways in which capital and the state produce natural landscapes, resources and sciences; human bodies, capacities and potentialities; and social infrastructures, spaces and cultures that the apparent silence referred to in the footnote appears.
My sense, as one who's worked with the theory a great deal - including Jim's problematic development of his own work, is that the thing he made least clear (perhaps because it's the hardest thing for Westerners - he and I included - to remember) is that environmental, feminist anti-racist and cultural historians have made it abundantly clear that the naturalization of landscapes and science, bodies and sexualities, spaces and cultures are all part of anti-historical anti-matieralist forms of modernism. Debates around the Anthropocene, queer theory and cultural geography all make clear that the naturalizations we inherit have been produced... and most of them quite recently, certainly only stabilizing over the last 200 years.
Despite how these comments might come across, I can't thank you enough for the effort to be fair to O'Connor and the second contradiction. It's rare to see a text work so hard to get things right.