Provincializing the anthropocene


back to issue

THE suggestion that we have entered a new geological era, the ‘Anthropocene’, an era in which humans for the first time must be counted as global agents, or drivers of change, cannot have escaped the attention of readers of Seminar. The assertion of a new form of agentive force for our species is subject to challenge in empirical terms, a point I discuss below. Evaluating the empirical sufficiency of the idea that significant human impact on the earth system is relatively recent is the subject of an ongoing research project to collate and commensurate historical, archaeological, and paleo-environmental evidence regarding the actual contours of the global human footprint (that is, a data-based rather than model-based reconstruction). While empirical sufficiency is important, the form that the Anthropocene debate takes is also of interest.

In this essay, I discuss the somewhat hidden Eurocentrism of the Anthropocene concept. To a surprising extent, the notion of an Anthropocene – and much of the analytical apparatus surrounding it – represents an effort to expand (rather homogenized) European historical experiences, frameworks and chronologies onto the rest of the world. I take the term Eurocentrism here literally, in that existing models tend to ‘build out’ from Europe and from the temperate zones, taking other regions as variants on an unmarked category. Building out from European history has given the Anthropocene discourse a particular flavour, not only within the scientific community but also among those who have embraced the concept with the fervour of the converted, chiefly humanists for whom the idea of global anthropogenic agency is particularly new and exciting, and hard scientists who have finally managed to naturalize human social relations into determinative models.

I argue here that the concept of the Anthropocene is unnecessary – not because humans have not changed the earth, but because we have done so throughout the Holocene. But even beyond this, it is important to note that the concept hides a disturbing extension of colonial discourse into a post-colonial world.


The title of this essay is of course a homage to Dipesh Chakrabarty’s Provincializing Europe, which seeks to dislodge European thought from the centre of the practice of history, using the study of South Asia as a vehicle for so doing.1 Here I would suggest not only that European historical experiences and the imagined relationships these imply about human population, land use, and human impact on the geosphere need to be decentred in analyses of anthropogenic environmental change, but further that the apparent novelty of a ‘geology of humans’ to both science and the humanities is just that – apparent. Provincializing the Anthropocene means not only that we no longer take European agricultural or industrial history as a starting point, or that we stop trying to project (and retrodict) proposed causal relationships between population and anthropogenic effects derived from a limited sample of human economic history, but also that we attend to the ways in which existing ‘western’ structures of thought and disciplinary practice overdetermine modes of agency – ‘human’ and ‘natural’.

Thus it is that those disciplines most enthusiastic about declaring an end to the Holocene, already the briefest geological period we know at ten thousand years, are those who, on the one hand, never before knew we were in it or, on the other, managed until now to analytically ignore or even erase human agency.


As parallel to Chakrabarty’s work, I offer here an alternative. Historical, paleoenvironnmental and archaeological research in India, among other places, shows us some of the limits of models and time markers built on an European base, challenging both the form and substance of work which directly feeds in to global and local climate models and, as such, to science, policy, and disciplinary imaginations of the human place in the world. The poser to this issue contends that history matters for environmental issues in the present, an assertion true in at least two senses. First, it is empirically true. A rising tide of research is showing that humans have, in fact, been both biological and even geological agents for a very long time; even the vast Amazonian rainforests once iconic of ‘pristine’ nature have been shown to be products of regrowth.2 This is a complex and variable history whose contours we must understand better, not only for their own sake, but for the present and the future.

History matters, too, in how we generate and understand evidence about human-environment inter-actions. Those of us in fields long dedicated to understanding such engagements know just how difficult it is to elude, for example, the fundamental nature-culture dichotomy that so pervades both thought and language. The Anthropocene debate, for all its empirical redundancy and European focus, may thus perhaps be in some ways a useful exercise after all. It has shown natural and physical scientists that humans can operate as more than simply ‘external’ disturbance factors to ‘natural’ processes, and humanists that they, too, may have a role to play in addressing the current environmental crisis.


Most proposals for an Anthropocene era adopt a rather limited historical perspective, assuming that significant environmental impact began only with the (European, and especially British) Industrial Revolution.3 This can become a self-fulfilling prophecy; consider the evidence on land transformation by humans reviewed in Ramankutty and Foley4 and Hook et al.5 which cover only the last 300 years. While the significance of recent anthropogenic change is beyond doubt, what is less clear is how novel such change really is. By shutting out consideration of longer-term change, we foreclose the possibility that anthropogenic change actually has a longer, more complex, or more variable trajectory than is generally assumed.

This issue is actually critical to the debate, since there is ample evidence to suggest that even in the absence of farming, humans sometimes drove vegetation change. For example, large-scale human burning has reshaped vegetation regimes from grasslands to prairies which were once thought to be entirely ‘natural’. Agriculture, of course, is another major means by which our species has reshaped not only vegetation, but also soils, slopes, hydrology, disease environments, the distribution of wild plants and animals and has made possible new configurations of human population.


Indeed, it is the onset of agriculture that provides another magic number in Anthropocene discourse, 6,000. Around six thousand years ago, farming came to Britain, Ireland, and northern Europe, initiating a new mode of subsistence that would have far-reaching implications. What is curious about the climate community’s interest in mid-Holocene transitions is not recognition of the significance of farming, however. It is the general acceptance of a date based on the rather late appearance of cultivation in what is arguably a small, remote, and unrepresentative part of the world to stand in more generally for the beginnings of agricultural impact. Elsewhere, farming is much earlier, more or less coincident with the onset of the Holocene, around 10,000 years ago.

In South America, Mesoamerica, Southwest Asia, South Asia, and East Asia, for example, we have ample archaeological evidence for early Holocene farming, a way of life that had significant implications for the non-human world. As farming diversified, some forms of cultivation, such as terraced hillsides and rice paddies reshaped landscapes in ways that rival those of a modern monocropped field. And many of these domesticated landscapes planted with rice, sugarcane, taro, and other crops are both widely distributed and temporally enduring. It is critical, therefore, to accurately assess the impact of not just 300, not just 6,000, but at least the last 10,000 years of human action on the earth.

The use of European and particularly northern European chronologies to periodize other parts of the globe is, of course, nothing new. India’s basic historical framework of Ancient, Medieval and Modern periods is but a modest renovation of the colonial Buddhist, Hindu, Muslim and British periods. In archaeology, too, terms such as Palaeolithic, Mesolithic, Neolithic, and Chalcolithic are European imports, categories whose movement across the globe with colonial science and subsequent naturalization have left a legacy of awkward constructs designed to paper over the fact that these terms do not always capture local realities effectively. If the starting points and inflection points of the Anthropocene enthusiasts are oddly northern European, so too are some of the scientific procedures, which built out, quite literally, from European experiences. I take this up in the following section.


Climate models are complex entities whose predictive power is built on understandings of causal and processual relationships, such as those connecting atmospheric conditions and temperature. Atmospheric circulation, ocean circulation, and land surface relationships are all important parts of climate models; land surface relationships include vegetation as one factor affecting heat, moisture, and albedo, among other things. Vegetation or land cover, the ‘living cloak’ of the earth, thus plays a role in climate. Current models suffer, however, from an inability to model anthropogenic land cover change,6 instead relying on simulations of climate-induced vegetation (‘potential natural vegetation’). We know, however, that human land use has been an important factor – or ‘driver’ – of change and that ‘potential natural vegetation’ has not always been the same as actual vegetation.

One way that this deficit has been addressed is through modelling. What are generally referred to as ALCC (anthropogenic land cover change) models posit relationships between historic population levels (themselves based on extremely rough estimates from historical data) and human induced land cover change. Models of past anthropogenic land cover change7 differ significantly from one another,8 so it is worth looking at how they operate.


Given the difficulties of aggregating and commensurating evidence about actual historical changes in land use and land cover – an effort now finally underway9 – ALCC models build from assumptions about the relationships between human population levels and their impact on vegetation. All models are simplifications, and my intention is not to critique the efforts of modellers who are, after all, making the best of a difficult situation. Still, the historical primacy of Europe finds resonance in the science itself as well as its temporal framing. The model developed by Jed Kaplan and colleagues,10 for example, was initially based on a simulation of change for European vegetation over the last 3,000 years. As they explain in a later article,11 ‘We expanded on this method in the current study by expanding the geographic scope to global and the entire time period from 8000 years ago to AD 1850, when the Industrial Revolution began to profoundly alter relationships between population and land use.’12


The use of algorithms based on temperate farming required the authors to introduce a tropical correction factor (a ‘potential productivity scaling’) in order to compensate for what they saw as unrealistically high anthropogenic land cover change in the tropics the model otherwise predicted. The model thus takes Europe as kind of baseline in terms of population-land cover relationships and ‘corrects’ for other regions. One can hardly blame scholars for building out from better-known to lesser-known instances, and the fact that the archaeology, paleoecology, and history of Western Europe is better studied and better synthesized than almost anywhere else on the planet is of course a different kind of reflection of Europe’s role in the world. Here science reflects the legacy of European power and its affluence, just as both chronological frameworks and watershed moments reflect a preoccupation with the specific history of Western Europe.


How much does this matter? It might matter a great deal. Western Europe, for all that it is impressively well studied, actually covers a modest portion of the world, around 7% of the earth’s land surface. The larger continents and the vast areas of the tropics are, in global terms, more significant, a weighting not highly evident in structures of scholarly attention or funding. Outside Western Europe (and within it, in places), industrialism came later and, in some areas, not at all. Further, global relations of extraction such as colonialism find little to no purchase in the theorization of land use and land cover changes within the modelling world. Indeed, as Malm and Hornborg argue,13 Anthropocene narratives that depict humans as a species ‘ascending to power over the rest of the Earth System,’ falsely naturalize intra-species inequality, noting that the creation of a fossil economy is more the product of particular social relations rather than of essential human biology.

This certainly holds true for farming as well. Not only is agriculture much older than 6,000 years in many regions outside Western Europe, but it was also often differently organized and sometimes more intensive. Intensively farmed landscapes such as the wet-rice systems of Asia and parts of Africa – along with the social and landscape transformations these entail – have been well documented by archaeologists and historians, systems that have in some places persisted for thousands of years. This deep legacy of environmental change has thus far failed to make much impression on Anthropocene enthusiasts, who persist in seeing only the last few centuries, and the future, as a time for the ‘geology of man’.14

Empirically, the creation of a new geological period seems superfluous. The key element of the Anthropocene – humans as agents of global change – is true of much of the Holocene as well. The ten thousand years of the Holocene is already a geological blip. Not coincidentally, it is also loosely conterminous with one of the most significant changes in human history, the domestication of plants and animals. We must come to terms with and better understand the anthropogenesis of the entire Holocene before we can evaluate the novelty or significance of present-day human impacts. Drawing a line at 1700, 1800, or 1850 runs the danger of implying – incorrectly – that older human-environment interactions were qualitatively different, perhaps in balance and harmony with nature and certainly with humans having had minimal impact on the natural world. Historical scholars already know this not to be true.


While there is much to critique in terms of the empirical substance of the argument for a new geological era, and more importantly, for a posited new relationship between humans and the earth system, there is also something disconcerting in a presumably global science so powerfully built out from European chronologies, histories, and modes of land use and vegetation. If postcolonial thinking requires that we provincialize Europe’s history and Europe’s knowledge systems, postcolonial global change will also call for a provincialization of the Anthropocene concept.



1. D. Chakrabarty, Provincializing Europe. Princeton University Press, Princeton, 2000.

2. S. Hecht, K.D. Morrison and C. Padoch (eds.), The Social Lives of Forests: Past, Present, and Future of Woodland Resurgence. University of Chicago Press, Chicago, 2014.

3. Jan Zalasiewicz, Mark Williams, Alan Smith, Tiffany L. Barry, Angela L. Coe, Paul R. Bown, Patrick Brenchley, et al., ‘Are We Now Living in the Anthropocene?’ GSA Today, 18(2), 2008, pp. 4-8. See Kidwell, 2015, for an overview of the debate over timing within the scientific community.

4. Navin Ramankutty and Jonathan A. Foley, ‘Characterizing Patterns of Global Land Use: An Analysis of Global Croplands Data’, Global Biogeochemical Cycles 12(4), 1998, pp. 667-685.

5. R. LeB. Hooke, J.F. Martin-Duque and J. Pedraza, ‘Land Transformation by Humans: A Review’, GSA Today, 22(12), December 2012, pp. 4-10.

6. G. Strandberg et al., ‘Regional Climate Model Simulations for Europe at 6 and 0.2 k BP: Sensitivity to Changes in Anthropogenic Deforestation’, Climate of the Past 10, 2014, pp. 661-680.

7. K. Klein Goldewijk, A. Beusen, G. van Drecht and M. de Vos, ‘The HYDE 3.1 Spatially Explicit Database of Human-Induced Global Land-Use Change Over the Past 12,000 Years’, Global Ecology and Biogeography 20, 2011, pp. 73-86. Also, J.O. Kaplan, K.M. Krumhardt and N. Zimmermann, ‘The Prehistoric and Preindustrial Deforestation of Europe’, Quaternary Science Reviews 28(27-28), 2009, pp. 3016-3034.

8. M.J. Gaillard et al., ‘Holocene Land-Cover Reconstructions for Studies on Land Cover-Climate Feedbacks’, Climate of the Past 6: 2010, pp. 483-499.

9. cover6k/intro

10. J.O. Kaplan et al., 2009, op. cit., fn 7.

11. This last point is attributed to J.O. Kaplan, K.M. Krumhardt, E.C. Ellis, W.F. Ruddiman, C. Lemmen and K.K. Golde-wijk, ‘Holocene Carbon Emissions as a Result of Anthropogenic Land Cover Change’, The Holocene 21(5), 2011, pp. 775-791.

12. E.C. Ellis and N. Ramankutty, ‘Putting People on the Map: Anthropogenic Biomes of the World’, Frontiers in Ecology and the Environment 6(8), 2008, pp. 439-447.

13. A. Malm and A. Hornborg, ‘The Geology of Mankind? A Critique of the Anthropocene Narrative’, The Anthropocene Review 1(1), April 2014, pp. 62-69.

14. P.J. Crutzen, ‘Geology of Mankind’, Nature 415, 2002, p. 23.