Driven both by the global loss of biodiversity and by the lack of knowledge about the vast majority of species that are being lost, conservation biologists and some of their allies in the environmental movement have called for and started a massive global census of biodiversity.1 Most prominently, E.O. Wilson has proposed a new mobilisation of scientific resources to complete a global survey of species.2 The identification of biodiversity ʻhotspotsʼ is the first step in a cascade of biodiversity investigation, Wilson hopes, which will culminate in a full inventory of global biodiversity and of the places where it is being lost. With complete information about the global population of biological species, Wilson is optimistic that we can undertake more refined conservation measures and ultimately move towards greater sustainability. In this article, I take the position that the global biodiversity census is as much about power and political life and the boundaries between nature and society as it is about scientific information gathering for conservationist ends. Drawing on Foucaultʼs concept of biopower, I suggest that the biodiversity census provokes us to think about the ways that biological nonhumans are embroiled in, and challenge, the technologies of power that see life itself as a political object. To the extent that the ʻactionʼ in the biodiversity census seems to rest largely with the human scientists who do the categorising, naming, counting, and labelling of nonhuman species, one analytic stance towards this scientific practice is an anthropocentric one. Here, the focus is on considering the field of social power in which scientific efforts take place, and asking questions about the discourses, resources and networks that make a biodiversity census plausible and possible. But what if nonhumans can be legitimately considered to be active participants in the field of biopower, just as human subjects who are censused are? Can nonhumans be sites of resistance to biopower, and disrupt its governing impulses? I argue that it is possible, and indeed necessary in the context of biodiversity, to extend the idea of biopower to include nonhumans as participants. Like human subjects, nonhumans are regulated and rationalised in matrices of knowledge and science, through which they are readied as productive resources for capitalism and mined as repositories of genetic information. Nonhuman participation in an ecological field of biopower also involves being part of an array of authority in environmental discourses, with effects for both humans and nonhumans, and constructing new possibilities for biosocial collectives. However, because nonhumans generally lack the capacity to be self-regulating subjects but are nonetheless necessary authorities in figuring biodiversity truth discourses, I suggest that they hold a different kind of place in biopower than self-regulating human subjects do. More specifically, because nonhumans constitutionally (rather than intentionally) refuse to internalise the meanings of human language, they are able to resist becoming self-regulating subjects to a significant extent, relying instead on their own semiotic interpretations of the environment and acting accordingly: for example, through migrating, reproduc- ing, consuming resources and filling ecological niches in unexpected ways, biotic nonhumans are constantly challenging the normalising will of biopower. At the same time, because environmental interventions to save species come to be justified on the grounds of global environmental well-being, the health and continued existence of nonhumans becomes an increasing imperative. In spite of the fact the biodiversity census may extend the reach of an ecologically unfriendly capitalism, I conclude that it will, on balance, reap important ecological goods in hybridising political practice and acknowledging extra-human locations of power. The article proceeds as follows. I start in Section 1 by reviewing the global biodiversity census proposal. In section 2, I consider how scientific power is extended through the census and what sources of power it draws on to do so. The science of censusing nonhumans requires a significant mobilisation of social power – financial, technological, institutional and discursive – in order to succeed. But because knowledge projects like demography have effects and techniques that reach beyond these sources of social power into life itself, I suggest that we cannot fully explain the importance of the biodiversity census through these means. In section 3, therefore, I extend Foucaultʼs concept of biopower into the nonhuman world as a means to explain the productive power and regulatory qualities of the census. Because the subjects of the biodiversity census are mostly nonhuman, however, I also consider how the concept of biopower mutates in light of their participation. 1. THE GLOBAL BIODIVERSITY CENSUS PROPOSAL Spurred by the problem of a major extinction event in which we do not even know what or how much is being lost, the basic knowledge-problem that the global biodiversity census is aimed at addressing is counting and describing all the species that currently exist in the world. According to the United Nations Environmental Programʼs (UNEP) Global Biodiversity Assessment, the best ecological estimates of extant species range from 3.5 million all the way to 111.5 million species, with the most likely figure at around 13.5 million.3 Only 1.75 million of those species are currently identified and described, however, or about 13 per cent.4 Each year, only 13,000 new species are formally described, a snailʼs pace given the magnitude of the task.5 Even when species have been described, the data often remains limited. Some species may have become extinct since being identified. May et al., for example, estimate that about 40 per cent of identified beetle species are known from a single examination in a single locality, sometimes an observation made decades ago.6 E.O. Wilsonʼs census proposal includes training and deploying a cadre of thousands of specialists in systematics, taxonomy and classification. He calculates that given 40 years of productive classification work per scientist, at the pace of ten species identified per year, approximately 25,000 professional lifetimes are needed, a number which ʻfalls well short of the number of enlisted men in the standing armed forces of Mongolia, not to mention the trade and retail personnel of Hinds County, Mississippiʼ.7 In the perfectly rational system that he hopes for, each expert would be assigned to a specific classificatory activity. While there are currently only three people in the world who are sufficiently expert in classifying termites, for example, Wilson would up their number proportionally to match the fact that termites comprise ten per cent of the total biomass of tropical regions. Wilson also champions investment in new computer technology that can combine scanning-electron microscopes with image-recognition software. Its goal would be to process and identify species instantly and to flag new specimens as they are passed through. The data held in the GenBank project, a computer database aimed at collecting information on all known DNA and RNA sequences, could be folded into this process. Given the massive numbers of species and the difficulty of resource mobilisation, other proposals suggest sampling procedures to get representative data on the global biopopulation. For example, Terry Erwin suggests that we aim for ʻmassive but achievable biotic inventoriesʼ that give us a relative fix on biodiversity.8 While it may not achieve Wilsonʼs goal of describing all species, targeting specific taxa and sampling certain species would have the effect of rationalising what is currently an ad hoc process. In the face of an ever larger human population, the United States census now makes limited use of representative sampling procedures, which its proponents argue make it more accurate than a large-scale but flawed collection of data about every individual.9 A global biodiversity census would aim to do the same. Some of the questions that surround the biodiversity census are scientific ones, such as concerns over the basic species concepts it employs and the problems of scale involved in identifying micro-species like bacteria.10 Given that the activities of science are not self-contained but always embedded in social relations, additional kinds of questions need to be asked, however. Yet little scholarly attention has been paid to what this biodiversity censusing effort means in political or social terms. Political ecologists have usefully inquired into the general effects of biodiversity discourses and the ways in which they are intertwined with regimes of power and governmentality,11 but they have not asked whether there is anything specific about the language and practice of censusing nonhuman bio-entities that is politically important. Environmental ethicists seem to have ignored the ethical dimensions of the topic altogether.12 Anthropologists and ethnobotanists have looked increasingly at the relationship between biological and cultural diversity, finding strong geographical and evolutionary correlations between the two and suggesting that a broader biocultural value linking nature and humans might be found in diversity complexes.13 However, they have not inquired into the importance of censusing as a technique by which the differences in biocultures might be constituted. Political scientists have written extensively and insightfully about the practice and effects of censusing human populations,14 but have not taken up how censusing nonhumans relates to political questions about power. The rest of the article aims to fill some of these gaps, particularly with respect to questions of how power functions in contemporary environmental science and politics. 2. THE BIODIVERSITY CENSUS AND THE SOCIAL POWER OF SCIENTISTS Taken as a socially-embedded activity, the production of successful scientific knowledge necessarily requires mobilisation of economic resources, expansion of institutional power, and discursive legitimation.15 Successful science, in other words, must be socially forged. This point does not suggest that the status of scientific truth-claims is fully dependent on social interests; rather, the point is that thinking about a scientific activity like the biodiversity census requires attending not only to activities of classification and arguments over species concepts, but also to how those activities are made possible in the first place. In short, what kind of social power does the biodiversity census draw on, depend on and reproduce? What makes the use of a biodiversity census seem intuitively obvious as the right tool to address biodiversity loss? Here, I consider two forms of social power, both crucial to the census project. First, I ask what financial and institutional resources are necessary to allow agenda-setting power and the capacity to steer future resources in advantageous ways.16 Second, I suggest the allure of the census rests partly on the way that biodiversity scientists are able to tap into discursive power, particularly the seductive power of discourses like panopticism and discovery.17 I consider these forms of power in turn. The institutional push for a global biodiversity census is centred in the United States, where it has harnessed major sources of funding, including the National Science Foundation (a $14 million fund for ʻplanetary biodiversity inventoriesʼ), the Packard Foundation and Harvard University. It has created a network of scientific-political organisations mainly based in the United States and Europe dedicated to censusing different parts of the natural world and promoting the ʻcompletion of the Linnaean enterpriseʼ18 into a ʻCatalogue of Lifeʼ – the Census of Marine Life, NatureServe, the Global Taxonomy Initiative of the Convention on Biodiversity, Species 2000 and the All-Species Foundation. There has been increased funding for the academic fields of taxonomy and bioinformatics.19 New professional lives have opened up around these resources – ʻeach species merits careers of scientific study and celebration by historians and poetsʼ, as Wilson puts it.20 The biodiversity census makes these possible first through the act of species identification and then through the subsequent study of species behaviour, ecological roles and potential uses for humanity. The key institutions of the global biodiversity census are organised around information and communications technologies, which worm their way right into the names and missions of the organisations involved. The Global Biodiversity Information Facility (GBIF) and the Integrated Taxonomic Information System (ITIS) (in partnership with US federal agencies including the Environmental Protection Agency (EPA) and the United States Department of Agriculture (USDA)) are two of the global clearinghouses for establishing quality specieslevel data, aiming to be ʻopen portalsʼ of biodiversity data.21 GBIFʼs mission is to ʻdevelop methods for sustainably using biodiversity … [by] rapidly, openly, and freely delivering primary data about biodiversity to everyone in the global community, using digital technologiesʼ.22 The political intent is universal access, while its method of delivery is technological. The universally-wired nature of ʻtheʼ global community is taken for granted, even as a global digital divide and the barriers of expertise suggest that no such community exists. One critical role that information technology plays in organising the global biodiversity census is in its ability to suggest a panoptic biological future.23 ʻImagine an electronic page for each species of organism on Earthʼ, Wilson asks us, ʻavailable everywhere by single access on commandʼ.24 Genealogically related to projects like Diderotʼs Encylopedie, the modern ʻEncyclopedia of Lifeʼ is the endpoint and ultimate goal of the censusing project, organised in a technology that claims to outrun space and time. The rhetoric of ʻachievingʼ a global biodiversity census also taps into complex Western narratives of discovery and conquest of nature (ironically, since the conservation agenda of the census is aimed in part at preserving the wildness of nature). This rhetoric also draws on the position of social power held by the modern sciences to reveal the unknown to human publics. Wilson exhorts supporters of the census to have ʻfaith in the sprint to the finish of the global censusʼ, promising that ʻunknown microorganisms ... will be revealedʼ and that ʻnever again, with fuller knowledge of such extent, need we overlook so many golden opportunities in the living world around usʼ.25 Similarly, the All-Species Foundation tells us that the global biodiversity census ʻoffers an unsurpassable adventure: the exploration of a little-known planetʼ.26 Finally, the discursive power of the census is connected to economic life, in the way that it renders nonhuman agents ready for postmodern capitalism as semiotic constructions (as in genetic codes for bioprospectors or images in nature videos).27 As Arturo Escobar argues, whereas ʻnatureʼ marked modern capitalismʼs attitude towards the nonhuman, ʻbiodiversityʼ is a term of postmodern capitalism, in the way that it readies nonhuman nature for semiotic use rather than material use.28 Indeed, postmodern capitalism may protect nature materially even as it commodifies it semiotically, as in the case of protecting the Amazon rainforest for its pharmaceutical potentials.29 Yet, as Escobar argues, ʻonce the semiotic conquest of nature is completed, the sustainable and rational use of the environment becomes imperativeʼ.30 That is, once biodiversity discourses help conserve an area as a biodiversity reserve which is made valuable in terms of code-commodity, it also becomes part of a political system of global environmental governance that continues to manage it for capitalism. Thus, conservation biologists have mobilised an expanding pool of financial and institutional resources, drawing in part on the seductive qualities of the dream of panopticism and the historical glory of exploration. To the extent that the agenda of global environmental governance is steered by their expertise, consensus and public statements,31 they have also garnered increased authority in speaking about matters of conservation, ecotourism and economic development. While these forms of social power (institutional, financial, technological, discursive) explain some of what is at stake in the biodiversity census, I want to suggest that the power of a global biodiversity census also rests in its hybridising force. It introduces nonhumans into the discursive heart of an otherwise anthropocentric modern human politics, economy and knowledge that has generally denied that nonhumans have formed a part of these projects.32 It creates a framework through which humans interact with, pattern and position the diversity of natural nonhumans. Understanding the importance of the biodiversity census therefore extends questions of power past its traditional human context into an ecological context. The important questions, then, are: for whom does this extended politics and power work?33 What happens to anthropocentric understandings of power upon the participation of nonhumans in the process? Does the (re)introduction of nonhumans tell us anything about the ʻwhereabouts of powerʼ, to use John Allenʼs phrase?34 Can power be nearly everywhere, or does it have a specifically local character in relations between things and thus an uneven distribution? To answer these questions, I turn to the idea of Foucaultian biopower, which considers how power works at the micro-levels of individual life, in relation to the more traditional forms of power considered thus far. 3. BIOPOWER AND THE CENSUS In this section, I start by considering how human censuses are understood to relate to politics and governing institutions, specifically through the Foucaultian idea of biopower. Because biopower is concerned with the ways that techniques like censusing operate on the terrain of ʻlife itselfʼ, it is particularly suited to thinking about the biodiversity census, which similarly involves a strategy for administering and rationalising life in ways that reach into nonhuman biological life as such.35 The consideration of censusing nonhuman life through the lens of biopower involves a basic trade-off. On one hand, the extension of biopower into nonhuman realms raises the spectre of a more subtle, but nonetheless corrosive, form of human power over the natural world. On the other hand, because power and resistance are always co-existent, nonhumans may disrupt the functioning of environmental governance in new and distinct ways. Specifically, biopower faces difficulty in creating self-regulating nonhuman subjects who internalise conditions of subjection. 3.1 Self-regulating subjects and the justification of power In contrast to absolute power commanded by the Hobbesian sovereign to ʻtake life or to let liveʼ, Michel Foucault argues that the modern form of biopower which replaced it in the nineteenth century was a regulatory and disciplinary form of power that involved ʻthe administration of bodies and the calculated management of lifeʼ.36 Biopower organised and administered life, through a variety of techniques or methods of power that dragged human life itself into the grid of power-knowledge. Institutions such as universities, public health agencies and the army, and regulatory forms of knowledge, such as demography and modern medicine, not only analysed life-processes but permeated them as well. Yet it was not just the use of these techniques or the presence of these institutions that characterised nineteenth-century biopower. Two parallel political shifts made biopower distinct from sovereign power. The first shift was that the right of the sovereign to have power over life and death was no longer justified based on protecting the sovereign from external threat (as in conscription in cases of war) or internal threat (as in the death penalty). Rather, the power over life and death was now justified in terms of the population itself, in modern democratic language. When war was waged, it was not to protect the sovereign, but in the name of the people and their continued existence. The justification for biopowerʼs interventions into the details of life – reproductive health, the ways in which we die, normalising vaccinations – was similarly made in the name of the population. One effect of this shift, Foucault notes, is that the death penalty became more difficult to sustain logically: ʻhow could power exercise its highest prerogatives by putting people to death, when its main role was to ensure, sustain, and multiply life, to put this life in order?ʼ37 In other words, the justifications for powerʼs activities have social effects that exceed the justifications themselves. What effect, then, will extending ʻprotecting lifeʼ to ʻprotecting biodiversityʼ have, if we consider the biodiversity census as an extended example of the logic of biopower? Part of the answer is that direct resource exploitation becomes more difficult, since power over life/death of animals, plants, insects or trees can no longer be justified by needs of the human population/sovereign to fight natural necessity with all its might.38 Ecological biopower thus involves both broader social trends like the rise of modern ecological consciousness and the emergence of conservationist ethics as part of it structural logic. At the same time, what becomes easier is both the management and regulation of nonhuman biological life by humans and the direct intervention in, and mutation of, biological and ecological life-processes, all in the name of bio- or eco-systemic ʻhealthʼ. Here, biopower can be understood as a logic of eco-governance that simultaneously subverts the resource-driven agenda of modern capital by trying to conserve material nature1 and enables and rationalises an entirely new form of intervention in life itself. The ecological sciences, on this reading, are one of the crucial institutions through which interventions into life are enacted, and the biodiversity census is one of its primary power-knowledge techniques. Thinking about biopower as involving nonhumans also has the consequence of changing the population in whose name powerʼs exercise is justified. If the idea of biopower adequately describes the intrusion of scientific, economic and regulatory techniques into the lives of nonhumans, then administering nonhuman life must be justified in the name of an expanded population as well – in this case, in the name of a global ecological population of species (and their genes), guided by an ethic of preserving and fostering biodiversity. In sum, in the name of planetary health (a metaphorical extension of modern biopowerʼs concern with human public health to a planetary scale), a new population is configured into which biopower intervenes, one explicitly composed of human and nonhuman members participating in ecosystemic communities. The second, parallel shift that Foucault notes was involved in the move from sovereign to regulatory power was that there was an unsettling effect on the practice of governing. Unlike sovereign power, regulatory biopower had the imperative to promote life, to ʻoptimise forces, aptitudes, and life in generalʼ, and its ʻhighest function was perhaps no longer to kill, but to invest life through and throughʼ.39 Yet, crucially, it had to do so in ways that did not make the population more difficult to govern.40 In other words, however life was politically managed, it had to be done in ways that ensured governability. This aspect of biopower was aimed at producing self-regulating subjects who internalised the qualities that promoted life but did not fundamentally disrupt social functioning. Self-regulating subjects were both efficient for power (since subjects did powerʼs work for it) and governable. Yet in this respect, the movement of nonhuman entities into the population in whose name biopower acts represents a potential location of freedom, or at least resistance, precisely because many nonhumans are constitutionally incapable of being self-regulating subjects who can internalise the conditions of subjection in biopowerʼs own terms. Nonhumans do not ʻknowʼ that they are a species or a member of a specific phylum, in those terms, or that they have a particular gene-sequence; rather, they have their own frameworks of understanding the moments of interaction with scientists and modes of environmental experience that guide their actions. In this respect, biotic nonhumans differ from the human subjects of biopower, who, as Foucaultʼs analysis suggests, become self-regulating subjects partly in virtue of the way that their consciousness is structured by biopower – by its language, its categories, and the techniques of self-making. When one looks at the minority of bio-entities that could be made partially self-regulating – genetically modified crops, pet clones, lab-grown tissue replacement, gene therapy and pre-selection, regulated game preserves in which species are fenced in or are trained to learn the boundaries of safe haven from human predators – they form a small fraction of the bio-population. Even in those cases, moreover, there are significant doubts over whether they can be made into self-regulating subjects in the same sense as humans. Their ability to accede to and internalise the normalising effects of power is limited by their biological constitution. Though they have varying kinds of subjective experience, they cannot be said to have the self-reflexive kind of subjectivity that humans do. Whereas human subjectivity is marked by the dual ability to be a distinct Self in contrast to the environment and to have the reflexive thought that ʻI am a Selfʼ, nonhumans are generally limited to the former possibility. If biopower cannot make most nonhumans into self-regulating subjects, then their governability rests solely on whether they can be controlled indirectly through the patterned grids of scientific prediction. Yet as groups and individuals, biological nonhumans routinely confound predictability, within their own kinds of subjectivities. They respond to ecological change by unexpectedly shifting migration patterns and locations. They expand in unanticipated ways into ecological niches that humans open directly (e.g., suburbs as feeding grounds for raccoons; rabbits introduced in Australia for hunting who subsequently overran the countryside; garbage dumps as sources of food for omnivorous bears) and indirectly (e.g., red-tailed deer population explosions in New England upon the over-hunting of deer-predators, causing substantial economic damage and fatal car accidents). Some species mutate at evolutionary speeds that far exceed those of humans (e.g., pesticide resistant strains of bugs or penicillin-resistant viruses). They sometimes form new relations with other species to the detriment of humans (e.g., birds as carriers of Asian bird flu). They remake ecosystems into new stabilities and undermine others. In short, while the lack of subjectivity and reflexivity in nonhuman populations is usually read as a source of acquiescence to human interrogation, it seems to also have an opposite effect, in that it constitutes them as ready sources of resistance and disruption to the desires of biopower to establish governable populations. Foucault says of biopower: ʻit is not that life has been totally integrated into techniques that govern and administer it; it constantly escapes themʼ.41 Nonhuman agents effect some of the very same escapes simply by living. Life itself escapes biopower. 3.2. Array of authorities A second component of biopower is ʻone or more truth discourses about the “vital” character of living human beings, and an array of authorities competent to speak that truthʼ.42 In the context of the nineteenth-century societies that Foucault was analysing, these truth discourses about living beings included fields like demography and medicine. In the context of twenty-first-century biopower, Paul Rabinow and Nikolas Rose suggest that they might be extended to include fields like genomics, cloning and reproduction.43 What is critical to the truth discourses surrounding these fields is that there be an array of authorities, like the human sciences, public health agencies or social theorists, that both problematise a certain issue and endeavor to intervene in the field raised by that problematisation. These authorities both render the issue socially visible and strive to rationalise solutions. Just as scientists play a central role as authorities in the truth discourses of human sciences, so too are they central to the truth discourses of biodiversity loss. The project of the biodiversity census involves conservation biologists as a critical part of the authorities competent to create a truth discourse around species loss and conservation, through the rationalisation of species into an ordered catalogue of nonhuman life. But the involvement of nonhumans in this truth discourse exceeds a simple presence as scientific objects, and raises questions about whether we can think of biopower as authoritative about nonhumans without any account of how nonhumans might themselves testify to those truth claims, resist them, or actively participate to some degree in the making of scientific knowledge. An alternative account of scientific practice that moves towards such a distributed model of authority comes from Bruno Latour and others involved in theorising an actor-network approach, in which human and nonhuman agents are seen as collaboratively involved in the construction of scientific truth claims.44 On this account, scientific authority depends in part on the (nonlinguistic) ʻtestimonyʼ of nonhumans who are marshalled by scientists to establish the veracity of scientific accounts.45 While it is still human scientists who problematise the field of biodiversity loss, then, the array of authorities competent to ʻspeakʼ the truth discourses surrounding that field is distributed among both human and nonhuman. This line of argument about authority has three consequences. First, it speaks to the question of the ʻwhereabouts of powerʼ. One of the criticisms of Foucaultʼs analytic of power is that once it is taken past a specific institutional site into the broader practices of governing, it seems to be everywhere and nowhere.46 In a broader sense, space and place themselves seem to disappear from power. In the case of biodiversity sciences, at least, extending the participants in the biopower formation to nonhumans gives power concrete locations – in the places where data is collected, in the laboratories where representative samples are brought under the microscope, and in the bodies of species who experience different life-possibilities and pathways because of the process of classification. What makes biodiversity discourses potent at all, in other words, is the marshalling of human and nonhuman authorities to its truth-claims in particular places and particular biological bodies. Second, the intrusion of nonhuman life into authority-generating processes like the biodiversity census disrupts the human subject at the centre of modern biopower by forcing a new set of constituents into the political field who cannot quite be captured by it in the same way. Because the biopower depends on the relative flourishing of life, biopower cannot speak and act authoritatively on behalf of the health of global biopopulations and simultaneously extinguish them. The necessity of nonhuman life for biopower both enables its extension and increases the living things that disrupt biopowerʼs desire for smooth governing. To return to Foucaultʼs analysis of the death penalty, if biopower (in contrast with sovereign power) complicated the state killing its own citizens, then so too should ecological biopower be seen to complicate the domination of nonhumans (in modern relations with nature), including the ongoing anthropogenic species extinction event. Finally, if the Latourian understanding of authority as distributed between both scientists and scientific objects is correct, this critique should also apply to humans, in the human sciences that Foucault was considering. The authority of scientific claims depends not only on scientists, academics and public servants, but on the very human subjects that make authoritative claims possible. Not only is resistance coexistent with power within human subjectivity, as Foucault claimed,47 but a more distributed kind of authority also resides in the practices of biopower, among the subjects who take part in its data collections, experiments and interventions. 3.3. Biosocial collectivities In Foucaultʼs historical analysis, the formation of biopower occurs within the context of the rise of the modern nation-state. Yet the biodiversity crisis, which I have been thinking of here in terms of an extension of human biopower into biodiversity-power, presents a political situation in which there is a veritable state of emergency (species loss), and yet there is no state in which to declare such a state of emergency. This observation is true in two senses. It is true, first, in a spatial sense, in that the biodiversity crisis is global, yet there is no global state in which such a crisis could effectively be addressed.48 While a layer of global environmental governance may be growing and even using environmental problems as a way through which its expansion is made plausible,49 it does not yet have the logical or practical means to resolve the wider problems of social justice and development involved in the biodiversity crisis.50 In a second sense, there is also no political formation that accepts the participation of nonhumans within its confines. The nation-state is a modern, secular and thoroughly human mode of organisation, one that is based around a community of humans who in turn decide what is right or good for themselves and their environment. Its reasoning is decidedly and openly anthropocentric, as ecologists who advance biodiversityʼs cause almost all accept as a necessary part of communicating the biodiversity crisis to human publics.51 Similarly, the discourses of global governance draw on a thin kind of global political community,52 but they do not grasp nonhumans as participants in their ideological vision. Global governance is hardly democratic with respect to humans, much less politically inclusive of nonhumans.53 Understood in this way, when global governance discourses address biodiversity loss, they do so either as the next logical step in the postcolonial mission (moving from ʻcivilisation, progress, poverty, [to] environmental sustainabilityʼ),54 or as simply another problem area for governance to address.55 In the context of biopower that I have been considering, what then is the effect of global biodiversity census without a global state? If the modern census was part of a power-knowledge formation that was both organised by and constitutive of the nation-state as part of biopower, then a global biodiversity census should have some transformative effects in constituting political forms. The hypothesis that I want to suggest is that a biodiversity census will help construct new ideas of a multi-layered and multi-species global community. As a way into this hypothesis, consider the effect of the modern human population census on ideas of community. Benedict Anderson, for example, argues that the modern census was integrally related to the creation of the categories necessary for the creation of postcolonial nation-states. Anderson argues that the ʻ(confusedly) classifying mind of the colonial stateʼ created identities through the census that might not have been recognised as such by those who were censused and classified.56 The census involved a ʻtotalising classificatory grid, which could be applied with endless flexibility to anything under the stateʼs real or contemplated controlʼ.57 Yet by undertaking this project, the conditions of postcolonial nationalism were shaped and forms of intelligibility were constructed (and imposed) that were not otherwise there. For present-day aspirants to statehood, a census remains an important marker of a consolidated national citizenry, as in the push for a Palestinian census as a way towards achieving a de facto Palestinian state.58 Similarly, then, a global biodiversity census might be understood as part of constructing a global biocitizenry and in forming a global ecopolitical community. Rabinow and Rose suggest that biopower must include ʻstrategies for intervention in the name of life or health, initially addressed to populations that may or may not be territorialised upon the nation, society or pre-given communities, but may also be specified in terms of emergent bio-social collectivitiesʼ.59 Like Foucault, Rabinow and Rose are concerned with human populations, but their use of ʻbio-social collectivitiesʼ that are not necessarily dependent on a territorial population suggests the possibility of forms of community that are not tied to the nation-state. If the nation-state is not necessarily the right analogy for biosocial collectivities of humans and nonhumans, then Michael Hardt and Antonio Negriʼs expansion of biopower past Foucaultʼs state-bound apparatuses of governing points towards how a de-territorialised collectivity might be theorised.60 If Foucaultʼs use of biopower was used in the concrete historical analysis of the transition in nineteenth-century Europe from the sovereign state of absolute power over life and death to one ʻin the name of the peopleʼ of disciplinary and regulatory power, Hardt and Negri push the historical analysis forward another step by drawing from Gilles Deleuze the idea of a transition from a Foucaultian disciplinary society to a society of control. Disciplinary society exists in relation to individual subjects, setting the parameters of what is normal and deviant, prescribing the rules of social behaviour, and constructing the boundaries of the social space in which its citizens rattle around. Power in disciplinary society is concentrated in institutions. By contrast, a society of control moves into the very interior of its subjects. It regulates not from outside but through its distribution and internalisation into ʻbrains and bodiesʼ.61 It is a form of power that exists in networked interiorities (i.e., the linked, self-regulating consciousness of subjects), not in external impositions, limits, sanctions or structurings. Power in a society of control is also unique in the way that it is able to make biopower its exclusive terrain of reference. For Hardt and Negri, it is also a more totalising form of power than disciplinary power – it ʻextends throughout the depths of the consciousness and bodies of the population – and at the same time across the entirety of social relationsʼ.62 It is such an organising power that Hardt and Negri see as globally operative in the social production of subjects. But Hardt and Negri are critical of the way in which the global society of control has been considered in a disembodied way. They argue that the abstractions of language, communication, and intellectual ideas have been given productive precedence over the material and corporeal. 63 In their neo-Marxist reading, the potential of a biopolitical analysis rests with its study at the level of labour, production and bodies. If Marxist analysis of modern capitalism understood communication as external or secondary to the material relations of production, Hardt and Negri want to read it as internal and immanent to production in postmodern capitalism. The semiotic reconfiguration of postmodern capital that Escobar suggests is thus the very productive activity in Hardt and Negriʼs framework, and biopower is a Foucaultian/Marxist hybrid.64 In such a framework, biodiversity is something produced, and the ʻthingsʼ of biodiversity – the individuals, the species, the communicative fabric of science around which knowledge of them is built, the development projects of which biodiversity is a component – are implicated in a global society of control. Like Foucaultʼs conceptualisation of power, Hardt and Negriʼs vision of biopower as a field or fabric of social and capitalist production is an image that makes us see a total matrix of power. Yet their relative exclusion of nonhumans is curious, since there is a slew of nonhuman agents outside of that productive field. There is a multitude of nonhumans, so to speak, that includes the ʻbacterial proletariatʼ, in E.O. Wilsonʼs colourful metaphor and the nomadic animal populations who routinely exceed national borders.