Collapse: How Societies Choose to Fail or Survive 
by Jared Diamond.
Allen Lane, 575 pp., £20, January 2005, 0 7139 9286 7
Show More
Show More

Are our dealings with nature sustainable? Can we expect world economic growth to continue for the foreseeable future? Should we be confident that our knowledge and skills will increase in ways that will lessen our reliance on nature despite our growing numbers and rising economic activity?

These questions have been debated for decades. If the debate has become increasingly shrill, it is because two opposing ways of looking at the world continue to shape it. If, on the one hand, we look at specific examples of natural assets (fresh water, ocean fisheries, the atmosphere as a carbon ‘sink’: ecosystems generally), there is convincing evidence that at the rate at which we currently exploit them, they are very likely to change character for the worse, with very little warning. On the other hand, if we study historical trends in the price of marketed resources, or improvements in life expectancy, or recorded growth in incomes in regions that are currently rich or on the way to becoming so, the scarcity of resources would appear not yet to have bitten us. If you were to point out that there are acute scarcities in the troubled nations of sub-Saharan Africa, those whose perspective is ecological will tell you that people living in the world’s poorest regions are poor because they face acute scarcities relative to their numbers; while those whose perspective is economic will argue that people experience scarcities precisely because they are poor.

You may think that a study of past societies could throw some light on the question. Most people have heard about the marked decline in the population of Easter Island some three centuries ago; they may know, too, of the pueblos abandoned by the Anasazi in the south-west United States eight hundred years ago, and of the disappearance of the Maya of the southern Yucatán a thousand years ago. They are dimly aware that these events had something to do with mismanagement of the local environment, reinforced in some cases by prolonged droughts. But the same people also know that the civilisations of China and India have evolved for about 5000 years and 3500 years respectively, and that neither is about to say goodbye. So what were the differences between the civilisations that disappeared and those that adapted more or less successfully to changing circumstances over the millennia? To say that the societies that have survived have done so because they managed their habitats well, maintained profitable relationships with their neighbours and prevented their members from killing one another, isn’t really to say anything: the survival of those societies itself proves all that. We have to know what the successful ones did in order to avoid the mistakes of those that went under. We would then want to know what those societies that have enjoyed material progress in recent centuries did that was not only not wrong, but may even have been right.

Jared Diamond, a professor of geography in California, looks for answers to these big questions by examining the societies that collapsed. The body of his book contains accounts of four such societies: the Easter Islanders; the Anasazi; the Mayans; and the Norse in Greenland, who died out six hundred years ago. Diamond demonstrates that the proximate cause of each collapse was ecological devastation brought about, broadly speaking, by one or more of the following: deforestation and habitat destruction; soil degradation (erosion, salinisation and fertility decline); water management problems; over-hunting; over-fishing; population growth; increased per capita impact on the environment; and the impact of exotic species on native species of plants and animals. As might be expected, the relative importance of these factors differs from case to case.

Diamond’s case studies have a thoroughness appropriate to the moral seriousness of his inquiry. He traces the way, step by tragic step, each of those long-dead societies, as their circumstances changed, made choices that led to their demise. In a moving account of the Norse in Greenland, for example, he shows that there was seemingly nothing wrong with their society. They had brought with them a devout and orderly set of practices relating them to God, to one another and to nature as they had known it in southern Norway. As Diamond discovers, the immigrants possessed an enormous amount of what today is known as ‘social capital’. And in that lay the source of their eventual collapse; for in their determination to maintain cultural coherence in a foreign environment, they made no attempt to learn from the native inhabitants – the Inuit – about the local ecology and the reasons the latter relied on fish and seals for sustenance. The immigrants’ culture had been built instead on cattle, crops and wooden houses. And in choosing to pursue their previous way of life in a fragile, alien landscape, they effectively signed their own death warrant. The Norse in fact died of starvation. This is history as geography, and it is deep and heady stuff.

In his search for the root cause of each collapse, Diamond deems no item of information innocuous, no evidence too awkward for consideration. He treats his raw materials like pieces in a jigsaw puzzle – written records, archaeological remains, oral history, palaeontological imprints – and places them against the broad canvas of those societies’ geography. He sifts through them to reach the most plausible reconstruction of what happened in the distant past. It is detective work of great skill and integrity. Some might wonder whether we need two hundred pages of analysis to reach an understanding of these collapses, but I find Diamond’s obsession with detail wholly understandable. When you embark on a project of such scope as this, the urge to dig and then poke around some more to check what else there might be is irresistible.

Diamond’s reading of the collapses is original, for nature doesn’t figure prominently in contemporary intellectual sensibilities. Economists, for example, have moved steadily away from seeing location as a determinant of human experience. Indeed, economic progress is seen as a release from location’s grip on our lives. Economists stress that investment and growth in knowledge have reduced transport costs over the centuries. They observe, too, the role of industrialisation in ironing out the effects on societies of geographical difference, such as differences in climate, soil quality, distance from navigable water and, concomitantly, local ecosystems. Modern theories of economic development dismiss geography as a negligible factor in progress. The term ‘globalisation’ is itself a sign that location per se doesn’t matter; which may be why contemporary societies are obsessed with cultural survival and are on the whole dismissive of our need to discover how to survive ecologically.

Diamond believes that the risk of collapses such as those experienced by earlier societies should be a matter for increasing concern today. To demonstrate why, he describes how Rwanda’s collapse as a society was brought about by unprecedented population growth in a subsistence economy operating in a fragile ecosystem. He also offers a natural history of China and Australia and the impact their populations are now having on their ecosystems. By his reckoning, the situation today is worse than it has ever been: as well as the old dangers of deforestation and so forth, we now have to deal with anthropogenic climate change, the accumulation of toxic chemicals, energy shortages and a near-full use of the Earth’s photosynthetic capacity. The concluding chapters of the book are devoted to speculations on the contemporary human condition, responses to dismissals of the concerns of environmentalists by sceptics, and a meditation on our hopes and the perils we face. Which is when the book skids and becomes a mess.

Diamond thinks that in order to demonstrate that mankind is currently engaged in unsustainable economic activity, it’s enough to offer a sample of the insults modern economies have been inflicting on nature. Thus he reports a case of deforestation here, increased pesticide run-off there, loss of biota somewhere else and carbon emissions everywhere. But we have been travelling that route for nearly five decades now: environmentalists have routinely pointed to the damage modern economic activity inflicts. Moreover, in recent years such environmental scientists as Paul Ehrlich, Edward Wilson and, most recently, Gretchen Daily, Harold Mooney and Walter Reid, have spoken out while taking far greater care with details and qualifications than Diamond appears to believe is necessary.

The more important reason why Diamond’s rhetoric doesn’t play well any longer is that it presents only one side of the balance-sheet: it ignores the human benefits that accompany environmental damage. You build a road, but that destroys part of the local ecosystem; there is both a cost and a benefit and you have to weigh them up. Diamond shows no sign of wanting to look at both sides of the ledger, and his responses to environmental sceptics take the form of ‘Yes, but . . .’ If someone were to point out that chemical fertilisers have increased food production dozens of times over, he would reply: ‘Yes, but they are a drain on fresh water, and what about all that phosphorus run-off?’ Diamond is like a swimmer who competes in a race using only one arm. ‘In caring for the health of our surroundings, just as of our bodies,’ he writes at one point, ‘it is cheaper and preferable to avoid getting sick than to try to cure illnesses after they have developed’ – which sounds wise, but is simply misleading bombast. Technology brings out the worst in him. At one point he claims that ‘all of our current problems are unintended negative consequences of our existing technology,’ to which I felt like shouting in exasperation that perhaps at some times, in some places, a few of the unintended consequences of our existing technology have been beneficial. Reading Diamond you would think our ancestors should all have remained hunter-gatherers in Africa, co-evolving with the native flora and fauna, and roaming the wilds in search of wild berries and the occasional piece of meat.

Here I should put my cards on the table. I am an economist who shares Diamond’s worries, but I think he has failed to grasp both the way in which information about particular states of affairs gets transmitted (however imperfectly) in modern decentralised economies – via economic signals such as prices, demand, product quality and migration – and the way increases in the scarcity of resources can itself act to spur innovations that ease those scarcities. Without a sympathetic understanding of economic mechanisms, it isn’t possible to offer advice on the interactions between nature and the human species.

Here is an example of what I mean. Forests loom large in Diamond’s case studies. As deforestation was the proximate cause of the Easter Islanders’ demise, he offers an extended, contrasting account of the way a deforested Japan succeeded, in the early 18th century, in averting total disaster by regenerating its forests. Now consider another island: England. Deforestation here began under the Romans, and by Elizabethan times the price of timber had begun to rise ominously. In the mid-18th century what people saw across the landscape in England wasn’t trees, but stone rows separating agricultural fields. The noted economic historian Brinley Thomas argued that it was because timber had become so scarce that a lengthy search began among inventors and tinkerers for an effective coal-based energy source. By Thomas’s reckoning, the defining moment of the Industrial Revolution should be located in 1784, when Henry Cort’s process for manufacturing iron was first successfully deployed. His analysis would suggest that England became the centre of the Industrial Revolution not because it had abundant energy but because it was running out of energy. France, in contrast, didn’t need to find a substitute energy source: it was covered in forests and therefore lost out. I’m not able to judge the plausibility of Thomas’s thesis – there would appear to be almost as many views about the origins, timing and location of the Industrial Revolution (granting there was one) as there are economic historians – but the point remains that scarcities lead individuals and societies to search for ways out, which often means discovering alternatives. Diamond is dismissive of the possibility of our finding such alternatives in the future because, as he would have it, we are about to come up against natural bottlenecks. We should be persuaded by the evidence that has been gathered over the years by environmental scientists that he is right, but simply telling us that we are about to hit bottlenecks won’t do, because environmental sceptics would reply that discovering alternatives is the way to avoid them.

If the future is translucent at best, what about studying the recent past to see how the human species has been doing? The question then arises: how should we recognise the trade-offs between a society’s present and future needs for goods and services? To put it another way, how should we conceptualise sustainable development? The Brundtland Commission Report of 1987 defined it as ‘development that meets the needs of the present without compromising the ability of future generations to meet their own needs’. In other words, sustainable development requires that each generation bequeath to its successor at least as large a productive base as it inherited. But how is a generation to judge whether it is leaving behind an adequate productive base for its successor?

An economy’s productive base consists of its capital assets and its institutions. Ecological economists have recently shown that the correct measure of that base is wealth. They have shown, too, that in estimating wealth, not only is the value of manufactured assets to be included (buildings, machinery, roads), but also ‘human’ capital (knowledge, skills health), natural capital (ecosystems, minerals, fossil fuels), and institutions (government, civil society, the rule of law). So development is sustainable as long as an economy’s wealth relative to its population is maintained over time. Adjusting for changes in population size, economic development should be viewed as growth in wealth, not growth in GNP.

There is a big difference between the two. It is possible to enumerate many circumstances in which a nation’s GNP (per capita) increases over a period of time even as its wealth (per capita) declines. In broad terms, those circumstances involve growing markets in certain classes of goods and services (natural-resource intensive products), concomitantly with an absence of markets and collective policies for natural capital (ecosystem services). As global environmental problems frequently percolate down to create additional stresses on the local resource bases of the world’s poorest people, GNP growth in rich countries can inflict a downward pressure on the wealth of the poor.

A state of affairs in which GNP increases while wealth declines can’t last for ever. An economy that eats into its productive base in order to raise current production cannot do so indefinitely. Eventually, GNP, too, would have to decline, unless policies were to change so that wealth began to accumulate. That’s why it can be hopelessly misleading to use GNP per head as an index of human well-being. Recently the World Bank published estimates of the depreciation of a number of natural resources at the national level. If you were to use those data (and deploy some low cunning) to estimate changes in wealth per capita, you would discover that even though GNP per capita has increased in the Indian subcontinent over the past three decades, wealth per capita has declined somewhat. The decline has occurred because, relative to population growth, investment in manufactured capital, knowledge and skills, and improvements in institutions, have not compensated for the decline of natural capital. You would find that in sub-Saharan Africa both GNP per capita and wealth per capita have declined. You would also confirm that in the world’s poorest regions (Africa and the Indian subcontinent), those that have experienced higher population growth have also decumulated wealth per capita at a faster rate. And, finally, you would learn that the economies of China and the OECD countries, in contrast, have grown both in terms of GNP per capita and wealth per capita. These regions have more than substituted for the decline in natural capital by accumulating other types of capital assets and improving institutions. It seems that during the past three decades the rich world has enjoyed sustainable development, while development in the poor world (barring China) has been unsustainable.

These are early days in the quantitative study of sustainable development. Even so, one can argue that estimates of wealth movements in recent history are biased. As regards natural capital, the World Bank has so far limited itself to taking into account the atmosphere as a ‘sink’ for carbon dioxide; minerals, oil and natural gas; and forests as a source of timber. Among the many types of natural capital whose depreciation has not been included are fresh water; soil; forests, wetlands, mangroves and coral reefs as providers of ecosystem services; and the atmosphere as a sink for such forms of pollution as particulates and nitrogen and sulphur oxides. If these missing items were to be included, the poor world’s economic performance over the past three decades, including China’s, would undoubtedly look a lot worse. The same would be true for the rich world.

There are further reasons for thinking that the estimates of wealth changes that I have been referring to are biased. They have to do with the way prices are estimated for valuing natural capital. Empirical studies by earth scientists have revealed that the capacity of natural systems to absorb disturbances is not unlimited. When their absorptive capacities reach their limit, natural systems are liable to collapse into unproductive states. Their recovery is costly, both in time and material resources. If the Gulf Stream were to shift direction or slow down on account of global warming, the change would to all intents and purposes be irreversible. We know that up to some unknown set of limits, knowledge, skills, institutions and manufactured capital can substitute for nature’s resources; meaning that even if an economy decumulated some of its natural capital, in quantity or quality, its wealth would increase if it invested sufficiently in other assets. The remarkable increase in agricultural productivity over the past two centuries is a case in point. But there are limits to substitutability: the costs of substitution have been known to increase in previously unknown ways as key resources are degraded. Global warming is a case in point. When the downside risks associated with such limits and thresholds are brought into estimates of sustainable development, the growth in wealth among the world’s wealthy nations will in all probability turn out to have been less than present estimates would suggest. It may even have been negative.

What I have sketched here is the correct way to determine whether contemporary economic development has been sustainable. It is also the correct way to evaluate public policy, for it tells me that a policy should be accepted if and only if it is expected to lead to an increase in wealth per capita. But you won’t find any of this in Diamond’s book. There is no evidence that he even realises he doesn’t have the equipment to hand with which to study our interactions with nature. Nor, as far as I can judge, has he tried to engage with his economist colleagues to learn whether ecological economists have anything to say on the matter. The many people who will be reading Diamond’s book will be fascinated by the historical case studies, but they will also be left with the impression that there is still no intellectual toolkit with which to deliberate over the most significant issue facing humanity today. Worse, they may not even notice they haven’t got the tools. So readers will continue as either environmentalists or environmental sceptics, each locked into their own perspective. It is a great pity.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 27 No. 12 · 23 June 2005

Partha Dasgupta assures LRB readers that the way many economists approach environmental problems is ‘the correct way to evaluate public policy … a policy should be accepted if and only if it is expected to lead to an increase in wealth per capita’ (LRB, 19 May). I am exasperated by such imperiousness, which will serve only to isolate economics further from the environmental debate. ‘There is no evidence’, Dasgupta continues, that Jared Diamond ‘even realises he doesn’t have the equipment to hand with which to study our interactions with nature’. He worries, too, that Diamond’s readers will ‘not even notice they haven’t got the tools’. The ‘tools’, ‘equipment’ and ‘toolkit’ to which Dasgupta refers are those of ‘modern economics’.

The vast majority of politicians, policy-makers and opinion-formers have no understanding or interest in intricate technical stories. More important, there’s little evidence that these technical stories have any influence on the thinking even of those who understand them. Market economists are supposed to respond to customer demand, not rebuke customers for failing to have the right toolkit.

Economists might begin by recognising the limitations of economic analysis. Dasgupta is right to emphasise the uncertainties that plague attempts to quantify natural capital, but wrong to imply that the solution is less ‘biased’ quantification. Since Keynes, most applied economists have accepted that genuinely uncertain outcomes cannot be quantified. Only the most austere Bayesians believe everything has a probability. Setting problems of uncertainty aside, quantification remains spurious. We would think it absurd to invoke cost-benefit analysis to justify say, homeland security policy. Environmental security should not have to pass such a test either. Bush and Blair know that cost-benefit analysis cannot legitimise their military expansion plans; they must appeal to morality. In this respect at least, environmentalism can learn from them.

Jonathan Aldred
Emmanuel College, Cambridge

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences