Forthcoming in D. Richardson

Department of Geography & Sustainable Communities, University of Wollongong and Geography, SEED, University of Manchester

Since Geography was founded as a university subject in the late 19^th century, geographers have defined their research and teaching with reference to nature, explicitly or implicitly. At times nature has been absolutely central to what most geographers do, at other times less so. After 1945, concerns about the credibility and utility of geographers’ pre-war inquiries led to a progressive split between human and physical geography, with the former largely losing interest in the biophysical environment and in questions of ‘human nature’. However, since the early 1990s a preoccupation with nature has, in a range of ways and for various reasons, become quite central to research and teaching across the discipline. This has not, however, led to a new unity among physical and human geographers. Whether this state of affairs is good or not is a matter of perspective. Today, three broad approaches to nature prevail in the discipline, each comprising multiple strands: a ‘traditional’ (or naturalist) one that regards nature as an independent object of analysis, a more recent one that questions nature’s naturalness, and a newer one that would dispense with the concept of nature (and its collateral terms, like wilderness) altogether. The second and third we can call ‘denaturalising’ approaches. This trio reflects geographers’ different conceptions of the character and aims of knowledge about the material world. The differences cannot – some might argue should not – be eliminated in the quest for an ostensibly ‘correct’ understanding of nature and its relations with society. If our knowledge of what we call ‘nature’ is profoundly social, then it follows that its validity cannot be justified with reference to ‘material realities’ alone. This is not, of course, to suggest that there is no biophysical world existing regardless of peoples’ diverse conceptions of it. After defining ‘nature’, the entry is organised chronologically and charts the changing ways geographers have interrogated it over the last century or more. It concludes by considering the relationship between the three just mentioned approaches.

‘Nature’, by definition, includes but also exceeds the meaning of ‘environment’. To consider how geographers have studied nature is, necessarily, to consider the discipline of Geography in its entirety (rather than just some of its sub-branches). Why so? First, in various ways the first university geographers saw their task as the study of human-environment relations – with humans perceived as more-or-less responsive (in a physical and psychological sense) to natural stimuli issuing from their immediate surroundings. For decades, academic geography was in some measure about ‘Earth writing’ in a literal sense: that is, the analysis of how the Earth materially inscribes itself on humans and vice versa. Second, even though Geography splintered into a largely nature-free ‘human geography’ and an environment-focussed ‘physical geography’ after 1945, nature necessarily served as a ‘shadow concept’ organising the former. In other words, by presuming that something called ‘society’ was different in kind from something called ‘nature’ (human and non-human), human geographers were able to justify a largely exclusive focus on the spatial organisation of people’s activities and creations. This focus was only possible by believing that the biophysical world could be ‘bracketed-out’ and left to physical geographers to study – making it what some theorists call a ‘constitutive outside’ or ‘absent presence’ in a semantic sense. Third, in recent years many human geographers have taken a deep interest in the more-than-human dimensions of the world, while others have concurrently challenged conventional understandings of ‘the social’ and ‘the human’. Though it has rarely involved collaborations with physical geographers, it has helped to dissolve the nature-society dualism that has for decades organised the academic division of labour in Geography. Fourth, also in recent years so-called ‘environmental geographers’, who position themselves between human and physical geography, have increased in number. In large part this is because of the ‘global environmental change’ research agenda now central to many funding agencies and intergovernmental organisations like the United Nations. But it also reflects the growing number of people who live in areas prone to perennial or periodic biophysical threats, like wildfires and floods. In short, whether our focus is on the discipline’s past or present, to write about the analysis of nature in Geography is to contemplate the nature of Geography. Despite this, there is no detailed history of how the two things have evolved over time (though see Castree [2005] for a sketch).

Since we cannot possibly cover such a large and complex terrain in a single entry, a parsimonious approach is required with all the risks of over-simplification this entails. We begin in the obvious place by defining nature and offering some wider considerations about its place in Western understandings of the world. We then highlight ways in which nature was a central object of description and explanation in the early decades of academic Geography, before summarising developments after 1945 when it became the almost exclusive preserve of physical geography’s various sub-branches. The fourth section focuses on the recent period in which, it is shown, ‘naturalist’ and ‘denaturalising’ approaches developed side-by-side in Geography, often without much contact.

Overall, we will discover that ‘nature’ provides an illuminating conceptual lens through which to view both the evolution of Geography and current fault-lines (analytical and political) in the discipline. It will be suggested that Geography has always had a ‘problem’ with nature. Disagreements over what is and is not natural, who should study it, and how, have, it will be shown, been flashpoints for the successive reconstitution of Geography as an academic enterprise since the discipline’s inception. Today, arguably, we are at a point where the ontological ‘obviousness’ of nature’s existence is so widely doubted that the nature of Geography may (again) be changing quite significantly. Far from ‘solving’ the problem, this ‘nature scepticism’ in its various current forms is symptomatic of ongoing struggles in Geography to capture the hearts and minds of practitioners, students and non-academic stake-holders of various stripes. It only constitutes ‘progress’ in how geographers think about nature seen from a certain viewpoint (whether it is one I share I will refrain from declaring here in the interests of impartiality). Accordingly, we give previous approaches their due rather than rushing past them to focus mostly on recent developments as if they eclipse all previous contributions. Most of these previous approaches share a philosophical commitment to nature’s naturalness (a commitment we will call ‘traditional’). As will be seen, more recent research aims to be post-traditional by questioning nature’s naturalness in various ways. Readers should note that there is a bias towards Anglophone research in what follows, and the analysis does not cover developments in continental Europe or elsewhere.

Nature is a very old world in the English language, and its meanings have varied in the detail through time. It refers to a wide range of phenomena in a plurality of different ways – this is why it is an unusually complicated word, some would argue the most complex of all. As a linguistic philosopher might say, it is a ‘signifier’ (word, symbol or sound) that has more than one ‘signified’ (a specific meaning) and these signifieds are attached to an astonishing number and range of material things (‘referents’), both human and non-human. Today, the term has four principal meanings which have endured through many decades. First, it designates the non-human world, especially those parts untouched or barely affected by humans (‘the natural environment’); second, the entire physical world, including humans as biological entities and products of evolutionary history (or, as some would have it, a deity); third, the power or force governing some or all living things (such as gravity or the conservation of energy); finally, the essential quality or defining property of something (e.g. it is natural for birds to fly and fish to swim). As a short-hand, we can (respectively) call these meanings ‘external nature’, ‘universal nature’, ‘superordinate nature’ and ‘intrinsic nature’ (see Figure 1)

Clearly, depending on the context of usage, the idea of nature can function as a noun, a verb, an adverb or an adjective; it can also be characterised as object or subject, passive or active. As geographer Neil Smith once put it, “Nature is material and … spiritual, it is given and made, pure and undefiled; nature is order and it is disorder, sublime and secular, dominated and victorious; it is a totality and series of parts, woman and object, organism and machine. Nature is the gift of God and … a product of its own evolution; it is a universal outside [human] history and also the product of history, accidental and designed, wilderness and garden” (1984: 11). In short, the word nature is – and has long been – promiscuous: not only polysemic and polyreferential, but utilised in a wide array of everyday and more specialised situations. We might therefore say it is a keyword rather than a buzzword.

Keywords, as cultural analyst Raymond Williams (1976) argued in his famous book of this name, have three characteristics. First, they are ‘ordinary’, which is to say used widely and frequently by all manner of people in all manner of contexts (private, commercial and civic). Second, they are enduring rather than ephemeral – they do not come-and-go in a way that buzzwords like ‘globalization’ or ‘post-modernism’ do. Finally, keywords possess what cultural critic Tony Bennett and colleagues, in their update of Williams’ book, call “social force” (2005: xxii). In other words, because their various meanings become normalized in a given culture they are able to govern (i.e. steer or direct) not only our thinking but also a wide range of practices resulting therefrom. A simple measure of the importance of ‘nature’ as a signifier is to imagine us dispensing with the term and its meanings altogether. The ‘hole’ in our language would be enormous. We would be rendered both inarticulate and incapable in large areas of our thought and action. In sum, if we did not already have the term in our present-day vocabulary, we would probably have to invent it.

This becomes even more obvious if we consider two other things. First, Williams rightly referred to “particular formations of meaning” (1976: 13, emphasis added) when explaining to readers how they might interpret a signifier like ‘nature’. His point was that in any given society at one historical moment there are likely to be family of keywords whose meanings bleed into, and borrow from, one another. We can say that the concept of nature is not exclusively associated with the word ‘nature’. Instead, the meanings are routinely signified by a range of other words that are (or have become) part of our collective vocabulary. In this sense, ‘nature’ is something of a “ghost that is rarely visible under its own name” (Olwig, 1996: 87). Its meanings often appear as collateral concepts (Earle et al., 1996: xvi). That is to say, they are signified by different keywords which refer us to similar or additional referents.

In the early 21^st century nature’s collateral terms include the following, among others: ‘environment’, ‘wilderness’, ‘gene’, ‘genius’, ‘biology’, ‘race’, ‘sex’, ‘biodiversity’, ‘animal’, ‘life’, ‘intelligence’, ‘human’, ‘instinct’, ‘blood’, ‘reality’, ‘climate change’, ‘mind’ and ‘ecosystem’. Some of these terms are relatively old, others relatively new. Some appear semantically simple and straightforward (though, in actuality, they are not), others more evidently complex. The precise ways in which they partake of the meanings of ‘nature’ varies, according to both the word and the context of reference. Most of them feature in discussions of both human and non-human nature, but some are used more exclusively. Most of them are also freighted with meanings that go beyond those connoted by the term ‘nature’. These collateral words are thus only partly – rather than exclusively – synonyms for the latter. It depends entirely on the circumstances of their invocation and usage (see Figure 2).
Figure 2 Nature and its collateral concepts (reproduced with permission from Castree, 2014: 18)
The principal meanings of the word nature (i.e. what it signifies) are routinely attached to all manner of material referents by way of other words (i.e. collateral terms). The words and meanings become conjoined in often complicated ways. However, these collateral terms may also signify meanings beyond the four signified by ‘nature’.

Second, circumstances aside, nature and its collateral concepts are members of a fairly select family of antinomies that can be said to structure Western thinking about the nature of reality. These antinomies include urban-rural, raw-cooked, wild-civilised, and authentic-artificial. Their meanings are best understood relationally rather than separately. Together they comprise a ‘semantic rule book’ that at some level governs specific acts of verbal and visual representation. They are the epistemic means by which Westerners have assured themselves that there is a measure of inherent order to the world. The binaries create boundaries and differences that are presumed to be hard-wired into reality, in part ‘by nature’, in part by design (see Figure 3).

If all this seems to complicate matters too much, there are some signals in the noise. Arguably, whenever we use the word nature and its proxy terms, we typically think we are making (or anchoring) ontological statements of a cognitive, moral or aesthetic kind. That is to say, we believe we are making (or vouchsafing) statements about a biophysical reality that exists independently of the words, concepts and terms we use to make linguistic sense of it. In this respect, we are apt to assume that ‘nature’ and its filial terms are ‘mimetic concepts’, that is ones whose meanings capture in words actually existing phenomena that exist ‘out there’ (or ‘in here’ if we are discussing our own physiology and neurology).

This is why I have devoted this section to discussing the concept of ‘nature’, rather than assuming that it is secondary to the material world it is intended to depict. Is this to suggest that ‘nature’ does not exist, only the concept and its proxy terms? In one sense, yes. This is not the same as arguing that the material things to which the concept refers do not exist: they assuredly do. But it is purely a matter of habit to call them ‘natural’ in the various ways and contexts that we choose to; indeed, it is arguably a matter of convention to divide the world into words and things, symbols and reality, mind and matter. For many (perhaps including some readers of this essay), this is no doubt a difficult argument to accept. It accords a lot of importance to language, its social origins and its practical effects; and it challenges the conventional idea that many or most concepts are semantic ‘mirrors’ that faithfully represent the material world. As noted above, nature, by definition, seems to be that which lies outside of and is irreducible to linguistic frameworks. But appearances can be deceptive. In the end, some (like Ingold and Williams) would argue there is nothing ‘natural’ about our habit of designating certain things as belonging to ‘nature’, nor about using those things as a reference-point to secure our ethical or aesthetic beliefs. Conversely, it follows that any new habits e.g. ones that would dispense with concepts like ‘nature’ and ‘society’ altogether, cannot simply be justified with reference to the ‘realities’ they supposedly respect. They too can be seen as contingent creations rather than necessities. This sort of ‘denaturalising’ approach to understanding nature has, as we will discover later, loomed large in human and environmental geography in recent years.

It is fair to say that the early university geographers took the ontological existence of nature as a given – it was considered ‘natural’ in one or more of the four senses identified above. Their concern was with phenomenal nature, rather than nature at the micro- or cosmic-scale. For academic Geography’s founders, the interesting question was the various ways that this nature affected different societies across the globe. From the get-go Geography was to be a ‘bridging subject’ devoted to intellectual synthesis. This was abundantly clear in a key 1887 address by the young Halford Mackinder to the Royal Geographical Society in London. Entitled ‘On the scope and methods of geography’, his lecture explained how and why Geography should take its place alongside other disciplines within the academic division of labour. His strategy, at once simple and audacious, was to call that division of labour into question. Geography, Mackinder argued, can “bridge one of the greatest of all gaps”: namely, that separating “the natural sciences and the study of humanity” (1887: 145). He was not alone in defining Geography as “the science whose main function is to trace the interaction of man (sic.) in society and so much of his environment as varies locally” (ibid.). At points east and west others were doing much the same, such as William Morris Davis in America, Paul Vidal de la Blache in France, and Friedrich Ratzel in Germany. The four men soon occupied important university positions, and were followed by similarly vigorous prosleytizers who quickly built-on the foundations their forebears had laid.

So began Geography’s career as a university subject and what historian of thought David Livingstone (1992: 177) called ‘the geographical experiment’. This was the attempt to keep nature and society under “one conceptual umbrella” in the face of “the incipient Balkanisation of knowledge that accompanied the professionaliz-ation of scientific specialities” (ibid.). There were two aspects to this. First, the likes of Mackinder, de la Blache, Ratzel and Davis saw the need to study nature as a whole not as a set of discrete parts. Where subjects like chemistry, physics and botany specialised in investigating select elements of the natural world, Geography would study all these elements in combination (as Alexander von Humboldt had famously sought to do early in the 19th century). This is what ‘physical geography’ was, according to earliest proponents like Mary Somerville. In her 1849 book of this name, she defined it as “a description of the Earth, the sea and the air, with their inhabitants the distribution of these beings, and the causes of their distribution” (p. 1). Her successors, influenced in part by Charles Darwin’s path-breaking theory of evolution by means of natural selection (1859), were sometimes given to discussions of ‘human nature’ and how it varied geographically in response to the conditions of regional physical environment. Second, this commitment to studying nature as an integrated, multifaceted system was accompanied by a desire to explore its two-way (‘vertical’) relationships with human societies in different places and regions (‘horizontal’ differentiation). For Mackinder and the other early geographers, it was important that nature be studied in context, as something that forms the basis of (and is affected) by human practices of an economic, cultural and political kind. In this sense, the study of human geography was to be deeply materialist with its roots – quite literally – in the soil. The resulting investigations of ‘human-environment’ relations could be conducted at a range of geographical scales, from the local right-up to the global.

Clearly, Geography had high ambitions in its fledging years as a university subject: it was, in terms of subject matter and scope, very much a ‘world discipline’. But it became increasingly clear that its ‘bridging’ aspirations were difficult to realise satisfactorily. The major problem was a practical one. Geography’s perspective on the world was so comprehensive, and its subject matter so compendious, that it proved very difficult to demonstrate causal connections between the component parts of the non-human world, let alone all these parts and various societies worldwide. Some, like Mackinder and Ratzel, had hoped that a geographical equivalent (or version) of the theory of evolution might be developed. But this hope was in vain. It was time consuming enough to provide mere descriptions of different societies and their physical environs, never mind plausible explanations. As a result, most early research publications by geographers were beset by what – with hindsight – were serious intellectual weaknesses. For instance, monographs that focussed on specific places or regions were often impressionistic by present-day standards, and filled with unverified speculations about how and why nature and human society were as they were in given situations. At worst, this shaded into what we would now regard as racism founded on supposed ‘natural differences’ between humans rooted in biophysical conditions and their evolutionary imprint. For instance, in the U.S. geographers like Ellen Semple and Ellsworth Huntington were apt to argue that certain physical environments produced human ‘races’ less intellectually or physically capable than Europeans. These arguments were presented as descriptions of ‘historical accidents’ that had blessed some races but not others. Indeed, both Semple and Huntington on occasion claimed that even ‘Europeans’ would ‘decline’ if they had evolved in a region like West Africa.

Such beliefs were consistent with certain versions of evolutionary theory – such as the work of Jean Baptiste Lamarck (1744-1829) – though certainly not with Darwin’s. With hindsight, they appear as academic attempts to justify the colonialism and clientalism – military, economic, politician and social – practised by West European countries and the USA in Africa and elsewhere. Through all this, nature was understood as both universal and external, intransigent and enabling, restrictive and productive, in ways that were ultimately contradictory. Where an apparently ‘hard’ environmental determinism was operative on some societies in some places, other societies elsewhere seemingly experienced only a ‘soft’ determinism. Here environmental conditions were said to produce either a ‘superior’ human nature or else ‘civilisations’ that had, in effect, built a rich house of ‘culture’ atop the inherent genetic and physiological universals of human biology. Intellectually, it was an approach that allowed Anglophone and continental European geographers to have their colonial cake and eat it too.
Post-1945 fragmentation: the emergence of a nature-free human geography, a environment-centred physical geography and a small disciplinary middle-ground

World War II was a turning point for Geography and its approach to the study of nature-society relations. Many who would subsequently gain positions in university geography departments served in the military between 1939 and 1945. The experience was formative for most, instilling a belief that precision, measurement and rationality were virtues to be aspired to. Aside from the skills of mapping, surveying and close field-observation, pre-war Geography struck many observers as offering precious little to the war effort. At the same time three other developments were significant. First, the discipline had failed to produce major books or intellectual innovations comparable to, say, John Maynard Keynes’ 1936 General theory of employment, interest and money. This became a cause for concern. Second, many outside Geography had successfully argued that the physical sciences and the social sciences (with the humanities) had to be different by virtue of their subject matter. There could be no overarching analysis of people and nature, it was argued, because the former possessed ontological properties quite different from rocks, rivers or ravines. For instance, humans are self-reflexive, linguistic, tool- making beings able to make their own history and geography – so argued philosopher Wilhelm Dilthey, among others. This argument drove a wedge between the two spheres – society and nature – that the original geographers had sought to bring together in a single intellectual frame. Finally, the embarrassment of ‘environmental determinism’ – the pre-war argument made by Semple, Huntington and others that some human ‘races’ were mere reflexes of climate, soils and resources – made some geographers determined to ‘raise their game’ intellectually.

After 1945 academic Geography progressively splintered into two major halves (human and physical), with each fragmenting into relatively discrete ‘systematic’ sub-disciplines. Consider some of the new studies of the natural environment published at the time. Bagnold’s (1941) The physics of blown sand and desert dunes inquired into physical process-phenomenal form connections in arid environments (Bagnold’s British military service had been in dryland regions). R. E. Horton (1945) used his engineering background to argue that the action of water over and through different types of soil and rock had consistent physical consequences that could be measured empirically – and even predicted. Finally, Strahler’s (1952) ‘Dynamic basis of geomorphology’ argued strongly that physical geographers should measure and explain how processes defined by universal laws create specific sorts of landforms given a certain set of ‘initial conditions’. Together, works like Bagnold’s, Horton’s and Strahler’s laid the ground-work for a physical geography where explanations were derived from the testing, by way of repeated observation and measurement, of refutable hypotheses. This was an altogether more specialised, more rigorous, less descriptive approach to the physical environment than almost anything found in pre-1939 geography. Within a decade key texts like Fluvial processes in geomorphology (Leopold, Wolman & Miller, 1964) were making this new kind of physical geography a serious proposition.

The turn to specialisation was undertaken in the hope that human and physical geography could become ‘spatial sciences’. They would discover – through careful measurement, experimental control (where possible), proposition testing and use of statistics – laws explaining spatial patterns (such as the common tendency of rivers to meander or migration flows to be inversely proportional to the size of destination cities). This implied a splitting of Geography’s subject matter into two, with intellectual unity (many hoped) maintained at the level of the perspective taken on the subject matter (rather than subject matter per se). Accordingly, human geography increasingly abstracted the analysis of political, economic, social and cultural practices from their biophysical integument. The pre-war fondness of some geographers for discussing ‘human nature’ (in the biological sense of mind and body) was also quickly abandoned. On the other side, physical geographers produced increasingly ‘scientific’ descriptions, explanations and even forecasts of Earth surface phenomena. Specialisation, new databases, new remote sensing capabilities and new computer technologies made this possible. But lingering aspirations to synthesis paid the price: physical geographers divided nature (in the sense of the planet’s physical environment) into the five areas that comprise the field to this day (namely, geomorphology, biogeography, climatology, hydrology and Quaternary environ-mental change. There was also a progressive move towards small-scale, short time horizon studies because hypothesis testing was very difficult for macro-scale analyses of (say) whole ecosystems or river basins. Where macro-scale physical geography persisted, it was often hived-off to other academic subjects – as happened with meteorology and climatology. Indeed, physical geography’s major post-1945 branches bled into cognate subjects – such as ecology in the case of biogeography. This made them increasingly interdisciplinary, and reflected their inability to police their own turf once the idea of a unified physical geography was progressively abandoned after 1945. Finally, all of the above meant that the study of human-environment relations became a minority pursuit, with ‘spatial analysis’ and the search for general laws, models and theories Geography’s new modus operandi.

These changes together bolstered geographers’ intellectual self-esteem and improved their external image within the world of higher learning. Ironically, though, Geography was effectively abandoning the study of human-environment relations at the very moment when the title of William Thomas’s 1956 edited book, Man’s role in changing the face of the earth, was becoming as obvious as it was profound. The Brit C.P. Snow’s famous complaint about the estrangement of ‘the two cultures’ – one literary-humanistic, the other scientific-rational – was (ironically) applicable to post-war Geography, the one subject that had made intellectual unity its raison d’etre. The long shadow cast by pre-1945 environmental determinism made many younger geographers wary of inhabiting the ‘middle ground’ between a newly scientific physical and human geography.

Notwithstanding all this, a number of geographers did try to occupy this terrain in ways that eluded the shadow. For instance, after the Second World War, many Western governments adopted a more hands-on approach to public welfare. This included a new desire to protect people from the effects of natural hazards, such as hurricanes, droughts and landslides. In this context, the American geographer Gilbert White pioneered an approach in which peoples’ perceptions of hazards became the major focus. In Human adjustment to floods (1945) and subsequent works, White argued that many individuals and communities living in high risk locations did not necessarily perceive themselves to be vulnerable, and so failed to take adequate measures to mitigate the effects of hazard events. In this way, peoples’ cognition was granted a degree of independence and flexibility rather than being assumed to bear the ‘objective’ imprint of their environs. Ian Burton and Robert Kates, among others, built on White’s approach and sought to identify different forms of ‘cognitive rationality’ specific to certain groups in certain hazardous locations (Burton, Kates & White, 1978). The presumption was that people’s perceptions of hazards – however inaccurate or distorted – could be typified and rationally explained, leading to tailored policy solutions that might better protect them and their livelihoods. This kind of geographical research was important in US and Canadian geography, with natural hazards researchers being directly involved in public policy agendas for hazardous regions in both countries.

Meanwhile, cultural ecology came to prominence by the early 1970s. Situated on the marchlands between cultural anthropology and cultural geography, this approach to society-nature study was pioneered by Julian Steward, Andrew Vayda, Roy Rappaport, Marvin Harris and Clifford Geertz. It was a critique of Anthropology’s mid-20^th century ‘culturalism’ – wherein cultural habits were thought to develop sui generis – and pre-war Geography’s muscular naturalism. Typically focussed on land- and water-based communities in the global South, cultural ecologists sought out those aspects of cultural belief and practice that seemed to be ‘functional’ adaptations to local ecology. In this way they re-materialised culture to its biophysical base, without succumbing to environmental determinism or racism/cultural chauvinism. The typical cultural ecologist would undertake detailed, long-term fieldwork into environmental usage and modification in one locality, as well as everyday and ritualistic cultural practices. The result was a set of holistic studies in which the metaphor of ‘homeostasis’ – borrowed from systems theory – loomed large, and in which ‘culture’ and ‘ecology’ were regarded as co-dependent, mutually adjusted, relatively stable, and internally complex domains of process, relationship and event.

Though cultural ecology was synthetic and hazards geography more specialised, they both prioritised immersion in local situations and eschewed the macro-focus of earlier human-environment geographers. Yet they also both, in their own way, held fast to the realism of their predecessors: ‘nature’ was taken as an independently existing domain that could be studied carefully to ascertain the threats, constraints and opportunities it afforded different people in different locations. This realism they shared with the newly scientific branches of physical geography. Though this conformed to the common-sense of the time, the cultural ecologists and hazards geographers’ efforts did not arrest the growing human-physical divide described above, and nor did they intersect with the Western environmental movement – which first gathered momentum in the late 1960s. By then it was clear to many that population increase, economic growth and mass consumption were having a profound effect on natural resource availability and the integrity of ecosystems. Mercury poisoning at Minamata Bay, Japan; the Torrey Canyon oil tanker spill; Rachel Carson’s (1962) best-selling account of how herbicides and pesticides got into the food chain: these and other events inspired the first Earth Day, the founding of Greenpeace and Friends of the Earth, and other seminal early-1970s environmental initiatives. Geography had a golden opportunity to make ‘human impact’ studies its main business – a possibility foreseen in 1956 in the earlier-mentioned Man’s role in changing the face of the Earth and foreshadowed a century earlier by geographer George Perkins Marsh (1864).

Quite why this opportunity was missed is hard to say. Although it was grasped at the teaching level, this was not really the case at the research level. Richard Chorley’s (1969) Water, earth and man – which called for a new focus on human-environment interactions – was arguably among the exceptions that proved the rule. Geographers conspicuously failed to analyse the anthropogenic local and global ‘environmental problems’ that became ever more apparent from the early 1960s. Morally, the discipline also virtually ignored the pro-nature (or eco-centric) arguments being made within the wider environmental movement. Instead, a relatively small number of geographers complemented the natural hazards work of White and associates with a rather anthropocentric focus on resource management. This resource analysis was usually empirical, quantitative and conducted in what geographer Tim O’Riordan (1976) called a ‘technocentric’ mode. In other words, this research looked at how best to conserve resources for present and future human needs. It rarely took issue with the fundamental causes of resource depletion and was very human-centred. Relatedly, a number of physical geographers were interested in the impact of human activities on the parts of the environment that interested them (and vice versa) (e.g. Hollis, 1975). Like resource management studies, their research had a policy dimension because environmental management needed to be based on a proper understanding of its objects (e.g. rivers, soil erosion, predator-prey relationships). Yet despite its apparent ‘ethical-neutrality’, this sort of research was arguably value-laden because it did little to challenge the institutions, human actions and rule-systems that generated environmental degradation in the first place. It was very much ‘status quo’ research.
The 1980s, 90s and early noughties: ‘re-naturalising’ human geography, de-naturalising an enlarged environmental geography, and finessing the nature of physical geography

By the mid-1970s, then, nature was a key object of analysis in Geography but in a different and more circumscribed way than a century earlier. A decade later things began to change. A significant development was that ‘critical human geographers’ – those politically on the Left of the subject – began to take a serious interest in not only nature but its collateral terms and referents too. Concurrently a new generation of environmental geographers (also critically disposed) began to focus hard on contemporary environmental problems, but in ways that challenged the status quo by de-emphasising the purported contribution of ‘nature’ (e.g. natural scarcity) to those problems. These developments, to be detailed momentarily, were at once unsurprising and novel. The radicalism reflected the formative influence of the worldwide anti-establishment protests that erupted in 1968 and lived-on for some years; but the ‘rediscovery of nature’ beyond physical geography was not, in fact, a delayed mirror image of the ‘green movement’ whose heyday was the 1975-85 period. As we will now see, many human and environmental geographers sought to de-objectify ‘nature’ and were relatively unconcerned about things like ‘environmental limits to growth’ or the ‘rights of nature’. In other words, and for reasons to be explained, they chose not to make Geography the discipline analysing and trying to reduce ‘the human impact’ on the non-human world (so continuing an existing trend). Their interest in nature took other forms and had other motivations.

Let us look first at developments in environmental geography, of which two are worth highlighting during this period: namely, the rise of (Third World) ‘political ecology’ and of ‘vulnerability’ approaches to understanding natural hazards. Cultural ecology, for the most part, adopted a resolutely local focus and tended to treat cultural groups and their biophysical milieu in isolation from the national and global scales. But this began to change from the mid-1980s onwards. The reasons were presaged in cultural ecologist Barney Nietschmann’s study of the Miskito Indians of the Nicaraguan coast. A field-trip in the early 1970s made Nietschmann aware that his chosen field area was beginning to be drawn into national and global commodity markets and was losing some of its former independence (Nietschmann, 1973). A decade later, Piers Blaikie and Harold Brookfield formalised this insight in the germinal books The political economy of soil erosion (Blaikie, 1985) and Land degradation and society (Blaikie and Brookfield, 1987). For them, cultural ecology’s focus on homeostasis and the relative autonomy of different cultures were increasingly being rendered unrealistic by two things: first, the growing reach of state power at the national level, and second the internationalisation of commodity production, distribution and consumption. ‘Political ecology’ was thus tasked with understanding how local resource use was being affected by wider social forces, and the accent was on asymmetries of power between ordinary people and the various actors (e.g. national states and multinational companies) affecting those peoples’ lives. A new generation of researchers, especially in North America, were inspired to uncover the complex chain of connections tying local land and water use decisions by (say) peasant farmers to global shifts in commodity prices, trade agreements and so on. Political ecology was thus ‘political’ in that it was critical of the actors and processes that were destabilising local land use practices. For instance, land users who degraded their local resources were typically seen in this approach as relatively blameless victims making difficult decisions in highly constrained circumstances explained by ‘external drivers’ beyond their immediate control.

If political ecology radicalised cultural ecology, and expanded its analytical horizons, so too did a new approach to natural hazards geography from the mid-1980s onwards. The studies of White and fellow-travellers tended, like cultural ecology, to bracket-off the local scale from its wider geographical context. They also abstracted people’s perceptions of hazards and the process of cognition from any broader consideration of social relations and social identity. This began to change when Kenneth Hewitt – a former adherent to the White approach – edited Interpretations of calamity (1983). A decade later, Ben Wisner and colleagues published the important book At risk: natural hazards, people’s vulnerability and hazards (1993). These texts had an international impact on Anglophone hazards geography. They suggested that ‘vulnerability’ to hazards was not simply a question of living in a naturally hazardous location: it was also a question of wealth, power, identity and social location. Drawing, like the political ecologists, on the radical ideas that flowered in the Western social sciences from the early 1970s onwards, Hewitt, Wisner and others argued that vulnerability was (i) variable in degree and kind within populations, and (ii) explained by the positions people occupied in regional or national social structures, and in world politics and economics more broadly. This kind of research directed policy makers’ attention away from peoples’ perceptions and conceptions of hazards (with associated public education programmes), and from technocratic policies like building flood walls to defend against storm surges. Instead, the attention was focussed on how poverty and powerlessness are socially produced, and on how vulnerable communities can be better protected against the problems triggered when hazard events occur, like a tsunami. Protective policies should, the argument went, be focussed as much on social welfare as on physical engineering solutions. Vulnerability was this seen as a societal problem whose downsides were starkly exposed when natural hazards (seen as ‘proximate’ not ‘root’ causes) occurred.

As these intellectual innovations were occurring, a disparate group of human geographers began to take an interest in nature. Their aim was not to understand its ‘real character’ and how it materially affected people. Instead, they sought to denaturalise understandings of nature and its collateral terms. Publications by scholars influenced by ‘post-structuralism’ and by Marxism respectively stand-out. The former were part of a wider ‘cultural turn’ in Anglophone human geography during this period, one that focussed on the origins and effects of various forms of social ‘discourse’. The latter were part of Marxism’s consolidation as a major approach in human geography. The denaturalising perspective that become de rigeur had three broad aims: first, to show that claims made about nature contained culturally specific assumptions about how the world is and ought to be; second, to show how references to nature were a means of furthering specific social agendas; and third, to show how these references actively concealed the two things just mentioned. Together these aims defined a ‘nature sceptical’ attitude in the specific sense that, while the concept of nature was taken to be real, so too the physical world it designated, the concept was seen to be ‘unnatural’, so too its role in societal discourse (lay and expert). It is this attitude that was flagged in the latter part of the section ‘What is nature?’ as an alternative to the presumption that ‘realism’ is the only sensible attitude towards those things we by convention call ‘natural phenomena’.

Out of many key contributions to the new ‘cultural studies of nature’ in 1990s human geography let me focus on just two authors. In a now classic essay and related book, Canadian geographer Bruce Braun (Willems-Braun 1997; Braun, 2002) applied post-structuralist Jacques Derrida’s ideas to the topic of nature and wedded them to post-colonial theory. The latter proposes that colonialism lives-on culturally even though it has largely disappeared economically and politically. Braun’s empirical focus was the early-1990s clash between environmentalists and a commercial logging company over whether to fell an area of ‘old growth’ temperate rainforest in British Columbia called Clayoquot Sound. The former saw Clayoquot as one of the last remaining spaces of ‘pristine nature’, while the latter regarded it as a valuable economic resource that should be logged in a responsible way for the good of the Canadian economy and those communities dependent on forestry jobs. In a detailed analysis of both sides’ representations of Clayoquot, Braun shows how the region’s ‘realities’ were made to appear quite different depending who was doing the looking. There was a clash between representations that made Clayoquot appear wild, intricate, threatened and special on the one side, and those (on the other side) that made it appear as one more ‘resource zone’ to be rationally harvested by hi-tech logging firms. In both cases, the authors of the representations claimed to be depicting Clayoquot as it actually was. But Braun’s point was that these represent-ations – which comprised books, pamphlets, press releases and newsletters – reflected the specific agendas of those promoting them. In other words, one could not adjudicate between them by testing their veracity against the ‘non-representational actualities’ of Clayoquot’s old growth trees.

Arresting though this insight was, Braun’s research contained a further surprise. Notwithstanding the differences in content and message between the environmentalists’ and forest company’s representations, Braun argued that they ultimately shared the same symbolic universe: a specifically Anglo-North American one that reflects the linguistic conventions and cultural assumptions of those colonists who spread through the US and Canada from the 17^th century onwards. He makes this point with reference to Clayoquot Sound’s small groups of remaining ‘native’ or indigenous peoples. These groups had, historically, lived a peripatetic existence and had used forest, lake, river and shore for generations to meet their material and symbolic needs. Yet both the environmentalists and the logging company fighting over Clayoquot’s future assumed that the region was largely empty. This, Braun argued, constituted a geographical expression of a specifically Western belief that nature and society are two separate things. Clayoquot’s indigenous peoples, he concluded, were thus victims of symbolic violence, even in the supposedly post-colonial conditions of modern Canada. Their history and present day claims to control of Clayoquot simply did not register in the unthinking assumptions made by the descendents of the original European colonisers. Even when seemingly ‘progressive’, Braun showed, references to nature can serve as politics by other means.

Around the same time as Braun was writing, Australian geographer Kay Anderson was reinterpreting one of nature’s oldest and most socially potent collateral concepts, ‘race’. In the 1980s ideas about race as biological differences between groups of homo sapiens made something of a reappearance in Western discourse, having been discredited in the 1940s and 50s for reasons we cannot go into here. In part this reappearance was linked to progress in the various sciences of human biology, which led to an international attempt to ‘map’ the human genome and an eventually abortive attempt to also map genomic diversity among humans. There was also, again courtesy of new scientific findings, a resurgent ‘nature versus nurture’ discussion in the West. Within the disciplines of human biology, physical anthropology, neuro-psychology and the young field of socio-biology, a debate has raged over whether people’s mental and physical capacities are mainly a function of genes and the like or a product of their socio-cultural environment. As this debate has unfolded, new biotechnologies have been invented that promise to alter ‘human nature’ so that behavioural ‘disorders’ or congenital diseases can, potentially, be engineered out of existence. In this context, many people worry that we are witnessing a new biological determinism to rival the eugenic beliefs popular in some Western countries in the 1920s and 30s. The concern is that beliefs about the supposed links between a person’s genes and their behaviour or appearance will be used to target those with a supposedly ‘inferior’ or ‘abnormal’ genetic constitution.

Writing in this context, Anderson explored how discourses about the supposed ‘actualities’ of a group’s ‘race’ were used to exclude them from the rest of a society both socially and spatially. In her 1991 book Vancouver’s Chinatown, Anderson showed how, from the late 19^th century, a majority white population of British descent in Vancouver stigmatized Chinese immigrants. She demonstrated how this majority fixated upon phenotypical and cultural differences between themselves and Chinese settlers, coding them in ‘racial’ terms and hierarchically too. This coding then served to hold the settlers at a distance in key arenas like employment and housing, reproducing European privilege. A later book on the origins of ideas about significant differences between human beings, Race and the crisis of humanism (Anderson, 2007), also focussed on the way culturally specific ideas are often conflated with the physical world they seek to make sense of. Anderson’s monograph explored the challenge Europeans faced in understanding aboriginal societies in Australian from the late 17^th century. Though holding Enlightenment concepts of humanity as a singular species and of all humans as biologically similar, new arrivals to Australia struggled to fit aboriginal people into their schema. Over time, Anderson shows, it was believed that these people were somehow closer to nature, almost animalistic, and thus not ‘human’ in the same way as Europeans were. Yet, far from being a ‘factual’ corrective to the Enlightenment idea of a ‘universal human nature’, Anderson demonstrates that this idea was itself culturally-infused with certain European perceptions of what it means to be ‘properly human’. The implication is that even contemporary ideas about human nature and intra-human group differences cannot escape the imprint of cultural preferences and dislikes.

Studies like Braun’s and Anderson’s paid careful attention to the origins, content and effects of various forms of ‘nature-talk’ and of representations of ‘natural phenomena’ more generally. However, they paid little attention to the material referents invoked, or to the political economic realm wherein people wrest goods and services from the Earth. The former oversight perhaps reflected a fear of reintroducing environmental (or biophysical) determinism by the back door. It underpinned publications focussing on how nature might be said to be a ‘social construction’ (see, for instance, Demeritt, 2002). By contrast, one of human geography’s leading Marxists, the late Neil Smith (1954-2012), focussed on both things. His influential book Uneven development (1984) argued that capitalism – the dominant system by which goods and services are today produced – generates geographically uneven development as a constitutive feature, not as a random occurrence or accident. This is because capitalism is inherently contradictory: its compulsion to increase wealth incessantly leads it to scour the earth for new investment opportunities, new markets, new raw materials, and new labour forces. According to Smith, the generation of profits and their deployment on a vast scale not only explains the rise and fall of cities, regions and whole countries; it also underpins what he called ‘the production of nature’. This, as Smith himself observed, is a counter-intuitive idea: “We are used to conceiving of nature as external to society … or else as a grand universal in which human beings are but small and simple cogs. But … our concepts have not caught up with reality. … [C]apitalism … ardently defies the … separation between nature and society, and with pride rather than shame” (1984: xiv).

What did Smith mean by ‘production’? Capitalism, Karl Marx argued, is an historically specific mode of making and exchanging commodities that began life in Europe during the early 18^th century. What makes it distinctive, he argued, is that commodities are produced not for their use value (i.e. their practical utility) but their exchange value (i.e. how much money they can command upon sale). Since they do not sell in order to get the same monetary sum back as it cost to produce their commodities, capitalists are fixated on ‘accumulation for accumulation’s sake’, employing workers to do their bidding and facing competitive pressures from rival capitalists. These pressures, Marx argued, necessarily force all capitalists to innovate and to seek a competitive-edge by making new products, hiring more-or-less workers, employing ‘smart machines’, relocating factories, and so on.

What has this got to do with ‘nature’? A great deal. According to Smith, capitalism has remade nature head-to-toe over the last two centuries, meaning that such things as genetically modified crops are only the latest in a very long line of material transformations of the non-human world. Nature has become a mere means to the end of profit-realisation, and in the process it has been physically reconstituted into an anthropogenic ‘second nature’. Though not formally referenced to his book, a brilliant 1991 monograph by historical geographer William Cronon illustrated Smith’s arguments in bracing detail. Nature’s metropolis tells the epic story of how the rise of Chicago during the 19^th century was connected umbilically to the formation of entirely new agricultural landscapes throughout the mid-West of the USA. For instance, in a superb chapter on grain production Cronon shows how money invested in new railways, new storage facilities and new farm equipment created a vast new geography of fields and fences that replaced the natural grasslands created by evolution and the actions of indigenous/native peoples. Capitalism, in both metaphorical and literal terms, revolutionised Chicago’s hinterland in a few short decades.

Smith’s arguments may seem somewhat overstated, even in an era where the genetic composition of species is being altered by science and technology. After all, even the ‘produced’ landscapes of modern, commercial farms are ‘natural’ in the sense that the animals, seeds and crops have not been created by people from scratch. So why did Smith not favour the use of ‘softer’ verbs (like ‘modification’ or ‘alteration’) when describing the capitalism-nature relationship? The answer is two-fold. First, he wanted to challenge those who continued to reference a non-social, supposedly pure, external ‘nature’ when advancing their own cognitive, moral or aesthetic arguments. His PhD thesis supervisor, David Harvey, had already set a precedent here back in 1974. Harvey strenuously resisted then-popular arguments claiming that the world was increasingly ‘overpopulated’, because there were more people alive than the earth’s natural resource could sustain. He argued that advocates of policies to limit birth-rates justified them by talking of ‘natural limits to growth’ that were supposedly fixed. This, Harvey argued, was an ‘ideological’ move because it neglected the fact that ‘optimum’ population numbers can only ever be defined relative to culturally specific assessments of what a ‘suitable’ standard of living is. Those neo-Malthusians arguing for population control, Harvey argued, were typically developed world inhabitants using spurious claims about ‘natural resource scarcity’ to imply that developing world poverty was caused by reckless procreation.

Secondly, Smith arguably favoured the metaphor of ‘production’ because he believed that in a capitalist world ‘nature’ and ‘society’ were no longer discrete domains of reality. In other words, they comprise an ontological unity not two ontological spheres. Terms like ‘alteration’ and ‘modification’ suggest some residuum of ‘natural nature’ untouched by capital. But Smith’s point was that those aspects of what we call ‘nature’ that are germane to our everyday lives are increasingly defined in relation to the needs and actions of capitalists. It’s not so much that all of nature is produced ‘all the way down’ physically – the Earth’s molten core, for instance, is hardly ‘produced’ by capital. It is more that so much of what we consider to be ‘modified’ or even ‘pristine’ nature is as it is because it serves the material and discursive demands of capitalist enterprises as they vary in time and space. And where it does not – as in the case of global warming, which is caused unintentionally by atmospheric pollution – this nature is the indirect result of the intentional production of other aspects of nature.

This said, Smith (ironically) said precious little about the material properties and causal efficacy of those things we by convention call ‘natural’, be they consciously ‘produced’ or not. His was a one-sided materialism that accorded considerable power to capitalism. It was left to others to fill this intellectual vacuum. There were two aspects to this. On the one side, some showed how capitalism was able to physically mould some parts of nature in the interests of profitability. Examples include environmental sociologist Jack Kloppenburg’s First the seed (1988) and David Goodman et al.’s From farming to biotechnology (1987). Both books focussed, like Cronon’s, on farming. They showed how science was used by agro-foods companies to break-down the physical ‘barriers’ posed by crops and farm animals to enhanced profits – barriers such as crops’ vulnerability to certain pests or a chicken’s inability to grow to adult size in a few weeks rather than months. In both books, the physical malleability of certain components of ‘nature’ was made plain in a way Smith only made theoretical mention of.

One the other side, other Marxist researchers interrogated cases where capital (or, rather, certain capitalists) is unable to impose its will entirely and where the agency of ‘nature’ becomes important. Two examples will suffice. Karen Bakker’s (2003) An uncooperative commodity showed how the physical properties of water made a huge difference to the way that privatising British water services unfolded post-1989. Its weight and bulkiness give it a certain intransigence, to which the new water market had to adjust. Likewise, Scott Prudham’s book Knock on wood (2005) showed how the Pacific coast forestry industry has, in key respects, had to adapt to the material challenges posed by softwood trees growing in mountainous environ-ments. Bakker and Prudham were not reintroducing a concept of non-social ‘nature’, of which Smith would surely have disapproved. Their point is that what we call ‘nature’ possesses a degree of agency and influence, but this is always defined relationally with respect to the changing needs and wants of capitalist firms not sui generis. It is thus never absolute and is thoroughly contingent and conditional.

The Marxian work on capitalism and nature prioritised economic processes and motivations. As we have seen, the suggestion was that the conventional idea of ‘nature’ has been rendered obsolete (in some or all of its four meanings) because biophysical entities can longer be understood in abstraction from the influences of capitalist enterprises. As Smith and Harvey also argued, this idea can be used for ideological purposes even as it has outlived its usefulness. This implies a gap between ‘discourse’ and ‘reality’, wherein the concept of ‘nature’ is used as a smokescreen by those with power or influence. This focus on the concept’s ‘performative effects’ is one they shared with the likes of Braun and Anderson.

Compared to all of the above, physical geography altered in quite conservative ways through the 1980s and 90s. It held firmly onto, while finessing, its commitment to the idea that (non-human) nature has objective properties that are amenable to impartial analysis. There were three noteworthy changes of emphasis. Firstly, the balance between pure and applied research arguably tilted slightly in the latter’s favour. According to Gardner (1996), this reflected a growth in the ‘environment industry’ after the first Earth Summit (in 1992) and, particularly, the field of environmental management. Issues such as desertification, water pollution, soil erosion and deforestation increasingly made it onto physical geographers’ research and teaching agendas (for more on this see Gregory, 2000: ch. 7). These geographers often sought to aid environmental managers by pin-pointing the physical changes caused by certain human actions (e.g. Burt et al. 1993). Secondly, at a more philosophical level, physical geography moved away from the ‘steady-state’ and ‘dynamic-equilibrium’ assumptions that had underpinned much 1970s and 80s research. Instead, practitioners began to appreciate that the environment is complex, often disorderly, and even chaotic in its operations. As Barbara Kennedy (1979) presciently noted in the late 70s, physical geographers are confronted with a ‘naughty world’ (see also Kennedy, 1994). This change in ontological assumptions was partly inspired by wider shifts in scientific thinking, notably the rise of ‘transcendental realism’, as well complexity and chaos theory (see Phillips, 1999). Thirdly, the rise of Quaternary studies and a new emphasis on ‘global environmental change’ meant that the study of environmental systems at large spatial and/or temporal scales underwent a revival. In a sense, physical geography’s historical origins as what Simpson (1963) called a ‘historical science’ were rediscovered, providing a counterbalance to the small-scale, process-form studies that had been so popular from the late 1950s. This meant that its credentials as an idiographic subject focussed on novelty and difference were reasserted, not at the expense of a nomothetic (i.e. law or pattern-finding) approach but as a recognition that general laws and processes can have non-general (unique) outcomes (as some ‘critical realists’ argue).

We have just seen that human geographers had ‘rediscovered’ nature by the turn of the new millennium by way of seemingly paradoxical ‘denaturalising’ approach. This approach, we have also seen, animated an expanded domain of environmental geography, leaving physical geography’s branches focussed on the non-human world and holding-fast to a realist ontology. What of the most recent period? How do geographers today understand ‘nature’ and what, therefore, is the current nature of Geography?

In human geography the de-naturalising sensibility already existing by 2000 has to some extent given way to a new approach that challenges the idea either that ‘nature’ is a socially constituted representation (or artefact) or that it is something existing regardless of human ideas and practices. This approach rejects the ontological distinction between society and nature that underpins most of the studies cited so far in this entry. It has roots in continental European philosophy and STS (science and technology studies): thinkers like Isabelle Stengers, Michel Callon, Donna Haraway and Bruno Latour have strongly influenced several erstwhile ‘human’ geographers since the mid-1990s. By the early 2000s two overlapping strands of research were already evident, both prevalent in British human geography: namely, one focussed on ‘hybrid geographies’ (closely associated with Sarah Whatmore’s [2002] book of that name) and one on ‘non-representational geographies’ (closely associated with Nigel Thrift’s [2007] Non-representational theory). Since we do not have the space to explore their subtle differences, we will simply highlight their commonalities.

Hybrid and non-representational perspectives maintain that dividing the world into ‘social’ and ‘natural’ entities simplifies reality to the point of misrepresenting it. The suggestion is that there is no neat, Maginot line that divides the world into two qualitatively different orders that somehow ‘come into contact’ with each other in various ways in different contexts. Instead, Whatmore, Thrift and like-minded geographers insist that we inhabit one reality not two, and that supposedly ‘natural’ things and ‘social’ things cannot be separated if we are to understand their character and effects. This chimes with Neil Smith’s arguments, but unlike Smith these geographers do not believe that capitalism is calling all the shots in our syncretic, mixed-up world. They further suggest that human engagements with the world far exceed acts of ‘representation’: they also involve touch, smell, hearing and physical interaction. We are multi-sensual actors, the argument goes, and what we are does not precede our relationships with all manner of other entities.

To flesh-out these arguments, there have been numerous recent studies of the almost endless ways in which so-called ‘social’ and ‘natural’ things are so entangled as to variously co-constitute, stabilise or alter one another (depending on the case). These studies are highly attentive to the fine details of different situations, and amount to a cognitive, moral and aesthetic call to attend to what the society-nature dualism has so long hidden from view. For example, in some of her work on wildlife conservation, Sarah Whatmore has argued that the seemingly unproblematic category ‘elephant’ needs to be called into question. Becoming an ‘elephant’ is, Whatmore and Lorraine Thorne (2000) show, a process that is contingent on the specific network of actors, institutions and physical environs in which individual pachyderms exist. There is thus, they argue, a notable set of differences between zoo elephants and ‘wild’ elephants, even though they are conventionally regarded as belonging to a single species possessed of stable and singular characteristics common to any creature so-named. Conservationists, they imply, could usefully pay more attention to the differentiated character of that which they’re seeking to conserve, and to the varied networks through which ‘becoming an elephant’ is achieved. To conserve ‘nature’ one needs to understand that it is not a discrete object or site to be protected, but a whole set of differentiated entities bound into complex relationships that might be very extensive in space and time. This obliges us to rethink both what is being ‘conserved’ and how in all nature conservation policies (see also Hinchliffe, 2008). More broadly, it means that any future politics has to be attentive to the relationships between, and admixtures of, what can no longer labelled as ‘social’ or ‘natural’ entities. This challenges most major extant political paradigms, be they environmentalist, socialist, liberal or any other.

Research in this vein has grown in size and diversity since 2000. It is now known, variously, as inquiry that has a ‘more-than-human’, ‘post-natural, or ‘post-human’ focus. Some of it examines the human body, yet without reintroducing fixed conceptions of ‘human nature’. It has also been called ‘neo-materialist’ because it shares physical geography’s deep interest in the properties and effects of materials, while challenging the atomist and reductionist assumptions about matter evident in some of its branches. A fine recent example of this expansively relational approach is Lesley Head, Jenny Atchison and Alison Gates’ (2012) Ingrained: a human biography of wheat.

However, this is not to suggest that earlier strands of denaturalising research by human geographers are passé. On the contrary, many studies continue to be published of who is speaking about ‘nature’ and its collateral terms, in what ways, with what intentions and with what effects. Many of these studies have used the analytical lens of ‘eco-governmentality’ (or ‘environmentality’) to make sense of discourses about everything from global climate change to ‘green consumption’. This lens derives from French philosopher-historian Michel Foucault’s notion of ‘governmentality’. Foucault argued that the process of governing people (and non-humans) exceeds the institutions of the modern state and is dispersed among multiple organizations and actors. He further suggested that government not only acts upon people but, as it were, through them by altering their sense of self and world. Governmentality as an analytical concept combines two aspects of governing: namely, ‘rationalities of government’ and ‘technologies of government’ (Miller and Rose, 2008: 15). The former specify the division of tasks and actions between authorities (e.g. political, spiritual, military, or familial) and express the ideals or principles that should direct government (e.g. freedom, justice, or entitlements). Such systems of thought also identify the nature of the objects to be governed (e.g. society, the nation, the population, animals, the economy) and characterise the persons over whom government is to be exercised (e.g. a flock to be led, legal subjects with rights, a resource to be exploited – see Rose and Miller, 1992: 178-179). As Lövbrand et al. (2009: 8) express it, “Whereas rationalities of government … render reality into the domain of thought, technologies of government translate thought into the domain of reality”.

How does this connect with issues of ‘nature’? In recent years three such issues have become ever more prominent in academia, politics, business and civil society. First, there are widespread concerns about the accelerating pace and heightened magnitude of anthropogenic climate change (‘nature as threat’). Second, there are equally widespread worries about natural resource exhaustion and anthropogenic degradation of terrestrial and marine ecosystems (‘nature under relentless pressure’). Finally, there is both excitement and concern about genetic modification, artificial life forms and other science-led attempts to intentionally remake the building blocks of life (‘nature de-natured’). In this context, what studies of ‘eco-governmentality’ do is trace the ways certain mainstream institutions are enrolling references to things like biodiversity, species extinction, resource conservation, genetic diseases, and carbon dioxide into new rationalities and technologies of government (see Rutherford, 2011). For instance, something as mundane and seemingly positive as promoting ‘ethical shopping’ can be seen as a contestable attempt by certain organisations (e.g. supermarkets) to govern the thoughts and actions of millions of people – in the process closing-off other possible ways of responding to global environmental change.

This kind of research, like that done earlier by Braun, Anderson and Smith, pays little attention to how the material world existing beyond rationalities and techniques of government might matter ‘in itself’. This may seem foolish in light of what many scientists tell us is the ‘dangerous’ magnitude of anthropogenic environmental change coming at us this century and beyond. Accordingly, some human geographers, and many environmental geographers, have been very focussed on the following issues: how the science of environmental change is communicated to and interpreted by ordinary people (e.g. Max Boykoff’s [2011] Who speaks for climate?); how advanced industrial societies can make a rapid socio-technical transition to a ‘low carbon’ state (e.g. Bailey and Wilson, 2009); how low income countries can adapt to the biophysical impacts of a warmer, wetter world (e.g. O’Brien & O’Keefe’s [2013] Managing adaptation to climate risk); how low income communities and vulnerable people can be made more resilient in the face of anthropogenic environmental change (e.g. Wilson, 2012); and how local alternatives to an environmentally destructive global capitalism can be fostered (Gibson-Graham & Roelvink, 2009). This is an incomplete list, but it speaks to the partial ‘renaturalisation’ of both human and environmental geography in light of the end, as some Earth scientists see it, of the Holocene (i.e. the 11500 year period of relative environmental stability coincident with homo sapiens’ species flourishing). I say partial because the kind of research just itemised generally does not take a humanly altered ‘nature’ as something whose character and impacts can be defined in absolute terms. In this sense, it carries forward the analytical sensibility first cultivated by the likes of Hewitt, Wisner, Blaikie and Brookfield many years ago.

However, some environmental geographers now argue for a more profound renaturalisation of geographical research and pedagogy. For instance, in his book Inhuman nature: sociable life on a dynamic planet, Nigel Clark (2011) claims that the strands of ‘denaturalising’ summarised earlier in this entry suffer a key weakness. In his view they tend to ignore – or to not take seriously enough – the very largest, inhuman forces operative in and on Earth (like ocean currents and volcanoes). The 20^th century, he argues, had fooled many Western geographers into believing that ‘high magnitude’ biophysical events are low frequency and spatially dispersed. Yet our climate-changed future might assault global humanity with many such events in time and space. In this light, denaturalising research – be it cultural studies of nature or ‘post-natural’ research – risks being irresponsibly ignorant of the corporeal threats and radical opportunities for a new human future attendant upon the Holocene’s end.

Clark’s belief in the autonomous existence and raw power of ‘big nature’ – albeit one imprinted by centuries of human activity – chimes with recent developments in parts of physical geography (and the wider environmental sciences). There many practitioners have regarded global environmental change as a further impetus to reclaim physical geography’s ‘big picture’, integrative focus (see Clifford, 2009). Many physical geographers have seen their research feature in the influential reports of the Intergovernmental Panel on Climate Change (IPCC), and a few are advocating the new idea of ‘planetary boundaries’, which humanity is said to compromise at its peril. Still others have, at times, lent their name to Earth System Science, a hoped-for super-science of the planetary environment that seeks to understand couplings, feedbacks and thresholds among the various sub-systems of land, air and water (see Wainwright, 2009). Through all this a rather traditional ontological realism prevails: global nature, one changing because of human impacts, is seen as amenable to analysis via remote sensing techniques, sophisticated computer packages, ground-truthing and logical analysis based on known laws. So it is that the long-standing nature-society dualism is preserved, despite the critical interventions of Whatmore, Thrift and others. Complementing this, a number of environmental geographers are keen to extend the scientific mind-set to both understand and better manage the ‘human impact’ on the world’s land-, water- and air-masses. Advocates of ‘land change science’ and ‘sustainability science’, like Billie Lee Turner II (Turner et al. 2004) and Robert Kates (2011) respectively, urge geographers to derive practical solutions to anthropogenic environmental problems by paying close attention to the scientific evidence. These geographers are, perhaps, less enamoured of the denaturalising impulses of many of their professional peers, and closer in spirit to most physical geographers. Though in no way neo-environmental determinists, they apparently accept the is-ought dualism that parallels the nature-society, object-subject ones discussed in the section ‘What is nature?’.

This said, there are exceptions. For instance, Mike Hulme – a physical geographer and former lead author of IPCC science reports – argues that the sciences of the Earth environment should be subject to socio-cultural analysis. This is especially true now that they may be assuming unprecedented importance in shaping public and political decision-making. In Why we disagree about climate change (2010) and related publications he challenges the idea that the environmental sciences deal merely in ‘facts’, leaving others to trade in debates about values and goals. More than this, though a scientist himself, he worries that the global claims made by the IPCC can too easily underpin idea of ‘global government’ and ‘global policy’ that are likely to alienate (even harm) many ordinary people whose lives are ineluctably local and situated. A new ‘climate determinism’, he suggests, may be afoot (see Hulme, 2011). If so, it comes on the heels of a neo-environmental determinism some detect in the writings of Geography’s most famous environmental analyst, Jared Diamond. His popular books Guns, germs and steel (1997) and Collapse: how societies choose to succeed or fail (2005) are part of a genre of ‘high-brow’ popular writing that recalls the grand analyses of Geography’s founding figures like Halford Mackinder. But they arguably suffer the same problems in their attempt to establish how the natural environment enables and constrains human flourishing. Gaps of data, questionable inferences, and contestable deductions are glossed in the search for historic patterns and cross-cultural trends in humanity’s relation to land, water and sky.

In light of all of the above, it should by now be clear that the various intellectual divisions internal to Geography will not be overcome in the name of a holistic analysis of our new environmental era, what some are calling the Anthropocene. Though some geographers hope for a reunification of the discipline (e.g. Herbert and Matthews, 2004), it is more realistic to expect greater interchange between certain physical, environmental and human geographers interested in some aspect/s of ‘nature’. This interchange has been interpreted as within-discipline experiments in, variously, ‘inter-’, ‘multi-’, ‘cross-’ and ‘trans-disciplinarity’ (see, for instance, Oughten & Bracken, 2009). Though some of these experiments are intended to combine the expertise of specialists in order to reveal the ‘total picture’, many are designed to foster mutual learning between diverse perspectives. Hulme (2010), for one, sees the latter as vital for both Geography and the wider society. For him, current debates about ‘nature’ – what it is, what it does, its value and its very existence – are not to be resolved. Instead, they are incitements to explore diverse views on the sort of world we believe we inhabit, and the sort of futures we might want to create. For Hulme, the ‘problem’ of how to understand what we call nature (and its collateral referents) is a productive opportunity not a hurdle to be overcome. It can and should spark democratic debate about the really big questions in our lives.
Conclusion

Despite its length, this entry has not represented the full range of work on nature by geographers past and present. Yet we have still covered an awful lot of ground. As we have seen, the centrality of ‘nature’ to Geographers’ inquiries has varied over the last century or more, but has never been marginal to their endeavours. In recent years it has captured the attention of physical, environmental and many human geographers, in contrast to the period 1945-80 when it largely fell to physical geographers to study and teach about it. This means nature has again assumed the wider disciplinary importance it enjoyed during Geography’s early years as a university subject – only now in very different, and more diverse, ways than the era of Mackinder, Ratzel, Semple, de la Blache, Davis and fellow-travellers. Today we can witness the influence of the accumulated history of geographers’ changing perspectives on nature (in any of its four main meanings and its myriad of referents). A traditional realist approach, albeit very different in the detail from a century ago, still persists, and looms large in contemporary physical, and parts of environmental, geography. Meanwhile, more recent two de-naturalising approaches exert influence in parts of both environmental and human geography – one focussed on representations (images, discourses, signs etc.) or material transformations (as in agriculture), the other on the dissolution of the supposed ontological divide between society and nature, human and non-human. Though we live in an era where many express alarm about the ‘realities’ of humanity’s impact on both non-human nature and its own corporeal nature, this has not led to a synthesis of the three approaches or the triumph of one over the others. Instead, intellectual diversity prevails.

Is this a weakness – perhaps a sign of incoherence, or a failure to respond responsibly to the ‘grand challenges’ of the early 21^st century? Some would say so. However, there is another view. By juxtaposing very different perspectives within a single disciplinary matrix, Geography today offers opportunities to combine and complicate forms of knowledge about the biophysical world that often exist in splendid isolation elsewhere in academic life. What can realists about ‘nature’ learn from ‘denaturalising’ critics, and vice versa? What difference might it make to research, public policy or scholar-activism? These questions challenge geographers and others to appreciate that, like all of the really crucial issues in life, there is no such thing as a ‘right answer’. Instead, there should always be struggles and disagreements over the way issues are framed and responses thereby delimited. Answers are thereby revisable, debate thereby stilled only for a period of years.