5
A Philosophy for Our Time

The experience with analysing science and technology not only led Latour to suggest an alternative social science. He used the conceptual apparatus that came out of his work in science and technology studies also to address the issues that increasingly troubled citizens, politicians, and scholars: ecological problems, the role of science in democracy, and ultimately the way we understand ourselves as modern people – or rather ‘postmoderns’, as some claim. Latour set out “to bring the emerging field of science studies to the attention of the literate public through the philosophy associated with this domain” (NBM: ix).

It's the late-1980s. In Eastern Europe, unrest spreads, which eventually will lead to the fall of the Berlin wall. Meanwhile, at international conferences, new global issues are discussed – the hole in the ozone layer, climate change, the depletion of the planet's natural resources and raw materials. The world becomes aware that it is facing new threats – threats, the causes of which transgress national borders, while their projected effects will affect future generations, and that thus exceed both the jurisdiction and the time-horizon of nation-based politics. Observing that environmental, economic and social threats increasingly tend to escape the established institutions for monitoring and protection, the German sociologist Beck (1986) announces that modern, industrial society has entered a new phase. He characterizes the transition as the advent of the ‘risk society’.

The new threats do raise questions about the role of science in democracy. Identification of the new threats depends on scientific knowledge. Citizens can detect stench themselves and their protests may urge politicians to take action; but that political action is required because exposure to some chemical compound may cause cancer in the decades ahead and because current levels of CO₂-emissions will eventually cause climate change with disastrous effects, is known only because of the work scientists have invested in the matter. So scientists no longer restrict themselves to what the idea of the role ‘value-free science’ in politics suggests, that is, to provide advice about the means to attain ends that democratic governments have decided to pursue. Increasingly their work defines the political agenda. However, the experts disagree on many issues, leaving citizens and politicians in the dark about what to think and what to do. How should a democracy deal with risks, uncertainty and the new role of science? The risk society calls for a “reinvention of politics,” Beck (1993; Beck, Giddens and Lash 1994: 1) argued.

Appraising Beck's social theory as “one of the most lively, creative and politically relevant forms of sociology developed in recent years,” Latour (2003) nevertheless thinks that Beck has misunderstood the problems we are confronted with. Beck thinks that the problems emerged because society has changed; according to Latour the crisis of modernity is even more profound: it shows that we cannot any longer live with our self-image as Moderns.

What is this self-image? For most of the twentieth century, the answer had seemed pretty clear. We – the peoples of Europe and North America – self-consciously declared ourselves to be Moderns. Modern, that is: to be no longer bound to the social practices and values of traditional societies, to be enlightened; in Kant's famous words: to be released from man's self-incurred tutelage. Science enables us to rationally control nature; democracy and the rule of law guarantee our political freedom. They are two of the keystones that define societies as ‘modern’ ones.

All diagnoses of modern society and culture are heirs to the analyses of Durkheim and Weber, the classical sociologists of the early twentieth century who tried to capture the nature of industrial societies that had developed in the century before. Weber, in particular, defined the terms in which for decades the debates about ‘modernity’ would be conducted.

Weber's extensive, comparative studies of European and non-European religions, economies and legal practices, led him to conclude that Europe and North America had developed distinctive features in three domains: culture, society and personality (Weber 1968 [1919]; 1972a [1922]; 1972b [1922]). For the domain of culture, he pointed to the development of modern science and technology, formal law, principles-based ethics and Western art, in particular harmonic chord music. For the societal domain, he cited the development of capitalism and rule-based, bureaucratic government. Finally, he characterized the modern West as having developed the personality of the professional, distinguished by its calculative attitude that allows strict separation between matters of business and personal concerns. This wide variety of social developments has a common denominator, Weber argued. All of them rely on methods, procedures, on calculability and predictability. What according to Weber makes the modern Occident exceptional in both time and space is its particular form of rationality, namely calculative, purposive-instrumental rationality, Zweckrationalität. He conceived Western rationalism as the product of a long process of intellectualization and rationalization.

Being modern means having learned to distinguish between ‘ought’ and ‘is’, between mind and matter, between on the one hand society and culture – the sphere of values – and on the other hand nature – the sphere of brute facts. Moderns know how to distinguish ends from means and they know that ends do not follow from facts, but are chosen on the basis of values. But, Weber stressed,

[t]he increasing intellectualization and rationalization do not […] indicate an increased and general knowledge of the conditions under which one lives. It means something else, namely, the knowledge or belief that if one but wished one could learn it at any time. Hence, it means that principally there are no mysterious incalculable forces that come into play, but rather that one can – in principle – master all things by calculation. This means that the world is disenchanted. One need no longer have recourse to magical means in order to master or implore the spirits, as did the savage, for whom such mysterious powers existed. Technical means and calculations perform the service. This above all is what intellectualization means.

(Weber 1968 [1919]: 593–4)

To understand and explain what set off the process of economic rationalization, Weber pointed to the sermons that helped to spread the asceticism of the Protestant work ethic (Weber 1972c [1922]). But “the most important fraction of the process of intellectualization” had started already earlier. Weber situates its start in Plato's allegory of the cave, in the discovery of the significance of knowledge and understanding that opened “the way for knowing and for teaching how to act rightly in life and, above all, how to act as a citizen of the state” and in the appearance of the rational experiment – “the means of reliably controlling experience” – during the Renaissance period (Weber 1968 [1919]: 596).

Rationalization has brought indubitable advantages, but Weber also acknowledged its downsides. As bureaucratic rationalism may become an “iron cage” from which there is no escape, it meant “loss of freedom”. Moreover, disenchantment of the world means “loss of meaning”. Swammerdam, a seventeenth-century Dutch scholar, could still perceive the proof of God's providence in the anatomy of a louse. But “[w]ho – aside from a few big children who are still found in the natural sciences – still believes that the findings of astronomy, biology, physics, or chemistry could teach us anything about the ‘meaning’ of the world?” Weber (1968 [1919]: 597) rhetorically asked. For rationalized people, the gods have deserted the world.

Weber's analyses gave sociologists and social philosophers a bone to chew on for many decades. Elaborating worries about ‘loss of meaning’ and ‘loss of freedom’, the Frankfurt School, up to Habermas (1981), turned Weber's concerns into an influential research programme. Identifying rationality and reason with enlightenment, the Frankfurt philosophers wondered why rationalism had resulted in capitalist exploitation, bureaucracy's iron cage, and ‘one-dimensional man’, rather than in freedom and emancipation. Having first declared reason to be an almighty god, they subsequently had to find an explanation for the evil that existed in rationalized societies. They had to address the secular version of the question of theodicy.

Weber's analysis began to meet more criticism only in the last decades of the twentieth century. Anthropologists showed other cultures to have intricate forms of rationality of their own. For example, trying to make sense of complicated legal procedures on Bali, Geertz (1983: 179) ironically observed: “we have here events, rules, politics, customs, beliefs, sentiments, symbols, procedures, and metaphysics put together in so unfamiliar and ingenious a way as to make any mere contrast of ‘is’ and ‘ought’ seem – how shall I put it? – primitive.” Kuhn's critique of the idea of a unified scientific method put the whole idea of an overarching concept of rationality guiding practices in question (Hollis and Lukes (eds.) 1982). If an appeal to rationality couldn't do much work for explaining scientific development, the prospects for the concept of rationality to perform the much more encompassing role Weber attributed to it were dim. Under the broad label of ‘postmodernism’, disappointed rationalists started to express scepticism about all ‘metanarratives’ – whether they related to politics, science, or the arts. But Weber is hard to surpass. As long as discussions about the nature of modern society take the concept of rationality as their battleground, they are conducted in his shadow.

That it is hard to get beyond Weber is little surprising. Indeed, it is undeniable that what he singled out as exceptional Occidental achievements do still characterize modern society: capitalism, science, formal law, bureaucracy, and the calculative attitude of modern professionals. So it may come as a surprise that Latour (NBM) claims: yes, indeed, we have all of that, but we have never been modern.

5.1 ‘We Have Never Been Modern’

Again, the analysis starts by calling attention to a text – this time not a text published in a scientific journal, but a newspaper article from Le Monde. Latour's daily reports that measurements taken above the Antarctic show the hole in the ozone layer to have grown ominously larger. Because the ozone layer prevents ultraviolet rays from harming plants and animals on Earth's surface, this is alarming news. A few lines later, the article shifts to CEOs of big companies who will modify their production lines to replace the chlorofluorocarbons that have been identified as the cause of the depletion of ozone in the stratosphere. The article continues to report that heads of state are involved in debates about chemistry, refrigerators, aerosols and inert gases too. However, the paper also says that meteorologists suggest that cyclical fluctuations rather than human activities account for the observed changes in the ozone layer. So both industry and governments no longer know what to do. Finally, towards the bottom of the page, the reader is informed that Third World countries and ecologists are talking about international treaties, moratoriums, the rights of future generations and the right to development. One and the same article

mixes chemical reactions with political reactions, links the most esoteric sciences and the most sordid politics, the most distant sky and some factory in the Lyon suburbs, dangers on a global scale and the impending local elections or the next board meeting. The horizons, the stakes, the time frames, the actors – none of these is commensurable, yet there they are, caught up in the same story.

(NBM: 1)

Expecting science, politics and economy to be clearly separated domains, with each of them having their own institutions, methods and procedures, any reader of Weber would be surprised to find a report in a serious newspaper that mixes facts and values, knowledge and doubt, human activities and ozone high up in the stratosphere, science, technology, politics and economy. But not Latour. Having been involved in science and technology studies by then for more than a decade, he is used to studying situations that do not fit the established system of disciplines that separates social issues from natural scientific ones, economic questions from political ones, and human activities from nonhuman factors. To analyse what went on in a laboratory in California, in Pasteur's Mémoire, and in late nineteenth-century developments in French agriculture and public health, or to answer the question who or what had killed Aramis, time and again the trick was to study the translations by which human and nonhuman actants form actor-networks that make up established facts, an ill-fated innovative public transport system, or turned Pasteur into a celebrity. So we might expect that the same trick will probably do also for understanding a news story in Le Monde.

In We Have Never Been Modern, Latour aims higher than merely repeating this message. Like Weber, he wants to understand what made the West exceptional and successful. And – going beyond Weber – he wants to understand how this success gave way to a world that today has to deal with an abundance of issues in which science and politics, technologies and ethics seem to have become tied together in Gordian knots. To address these concerns, Latour argues, another dimension of analysis has to be added. Science and technology studies have brought only half of the story. The leading hypothesis of We Have Never Been Modern is that

the word ‘modern’ designates two sets of entirely different practices which must remain distinct if they are to remain effective, but have recently begun to be confused. The first set of practices, by ‘translation’, creates mixtures between entirely new types of beings, hybrids of nature and culture. The second, by ‘purification’, creates two entirely distinct ontological zones: that of human beings on the one hand; that of nonhumans on the other. Without the first set, the practices of purification would be fruitless or pointless. Without the second, the work of translation would be slowed down, limited, or even ruled out. The first set corresponds to what [Latour has called] networks; the second to what [he] call[s] the modern critical stance. The first for example, would link in one continuous chain the chemistry of the upper atmosphere, scientific and industrial strategies, the preoccupations of heads of state, the anxieties of ecologists; the second would establish a partition between a natural world that has always been there, a society with predictable and stable interests and stakes, and a discourse that is independent of both reference and society.

(NBM: 10–11)

So we have a triple task: we have to study the practice of translation, the practice of purification and the relation between the two. The first task has been taken up by science and technology studies. Weber had already started the second one. Purification, that is, distinguishing between on the one hand the sphere of human values and interests and on the other the disenchanted sphere of the inanimate world, is closely related to what Weber called rationalization. However, there are two crucial differences. Latour claims that purification is a practice, not an acquired attitude; purification requires work. And with regard to the relation between these two practices, he claims that the practice of purification comes second to the practice of translation.

If purification comes second to translation, one should be able to study its emergence out of translation work. In a long discussion of Shapin and Schaffer's (1985) brilliant study Leviathan and the Air-Pump, Latour argues that the authors had done just that (Latour 1990a; NBM: Chapter 2). Explicitly intended as an exercise in the sociology of scientific knowledge (Shapin and Schaffer 1985: 15), Leviathan and the Air-Pump reconstructs and analyses a debate between Boyle and Hobbes in the 1660s and early 1670s.

Boyle and Hobbes agreed on almost everything. They both “want a king, a parliament, a docile and unified church, and both were fervent subscribers to mechanistic philosophy” (NBM: 17). Their dispute concerned the legitimacy of knowledge produced on the basis of experiments. Boyle claimed that his experiments, overseen by reliable gentleman-witnesses, produced authenticated facts. Hobbes demanded the certainty of mathematical proof, the kind of knowledge that can be checked not only by a few gentleman friends gathered in a private space, but publicly, step by step, by anybody. But behind the epistemological issues, big political questions about social order loomed for which Boyle and Hobbes offered rivalling solutions.

One solution (Boyle's) was to set the house of natural philosophy in order by remedying its divisions and by withdrawing it from contentious links with civic philosophy. Thus repaired, the community of natural philosophers could establish its legitimacy in Restoration culture more effective to guaranteeing order and right religion in society. Another solution (Hobbes's) demanded that order was only ensured by erecting a demonstrative philosophy that allowed no boundaries between the natural, the human, and the social, and which allowed for no dissent within it.

(Shapin and Schaffer 1985: 21)

To explain why Boyle's solution prevailed, Shapin and Schaffer proceed in three steps. Displaying in fascinating detail the conventions and the crafts the two opponents required for producing authenticated knowledge, and by laying out the new style of reporting and discussing facts Boyle introduced and the proliferation of scientific instruments, Shapin and Schaffer argue that the solution to the (epistemological) problem of knowledge involved laying down the rules and conventions of the intellectual community, with Boyle and Hobbes opting each for a different form of intellectual life. Secondly, Shapin and Schaffer argue that the knowledge thus produced became an element in political action in the wider society. Finally they claim that “the contest among alternative forms of life and their characteristic forms of intellectual product depends upon the political success of the various candidates in insinuating themselves into the activities of other institutions and other interest groups. He who has the most, and the most powerful, allies wins” (Shapin and Schaffer 1985: 342).

As one of the founders of the Royal Society and throughout his life being one of its most notable fellows, Boyle had the better hand. He could mobilize powerful friends. Experimental science was instituted as a source of authenticated knowledge, separated from politics, state and religion. Already in 1663, drafting the constitution of the Royal Society, Hooke articulated the divide between, on the one hand, the business of science and, on the other hand, politics, religion and morality:

The Business and Design of the Royal Society is – To improve the knowledge of natural things, and all useful Arts, Manufactures, Mechanick practices, Engynes and Inventions by Experiments – (not meddling with Divinity, Metaphysics, Moralls, Politicks, Grammar, Rhetorick, or Logick).

(cited in Ornstein 1963 [1913]: 108 n. 63)

Boyle won. However, in the last sentences of their book, Shapin and Schaffer claim: “[a]s we come to recognize the conventional and artefactual status of our forms of knowing, we put ourselves in a position to realize that it is ourselves and not reality that is responsible for what we know. Knowledge, as much as the state, is the product of human actions. [So, Boyle may have won the dispute, but] Hobbes was right” (Shapin and Schaffer 1985: 344).

Given his former debates with Schaffer and other sociologists of scientific knowledge, we cannot expect Latour flatly to accept this conclusion. Shapin and Schaffer's claim to have explained the authentication of the content of experimental knowledge by political action in the context of British society fails according to Latour, because “neither existed in this new way before Boyle and Hobbes reached their respective goals and settled their differences” (NBM: 16).

So, much to the chagrin of the authors, Latour rewrote their argument, cleverly turning Leviathan and the Air-Pump into another kind of study, namely a comparative anthropological investigation of the dual process of establishing science and politics as separate domains, rather than an exercise in the sociology of scientific knowledge. On Latour's reading, what made Boyle prevail and Hobbes lose the dispute was something else than Boyle having powerful gentleman allies.

Boyle had acquired an airpump (invented by Von Guericke in the 1650s) and he claimed as a matter of fact that with this machine he had created a vacuum in a glass globe (the airpump's receiver). This claim was unacceptable for Hobbes, who adhered to ‘plenism’, a long standing philosophical doctrine to which no genuine potentiality of being can remain unfulfilled and that, as a consequence, conceived a vacuum to be non-existent and even unthinkable. However, as Shapin and Schaffer had noted, Hobbes's opposition was not only based on philosophical grounds. He was driven by fear of disruption of social order, the same fear that had motivated him to write his famous treatise on political philosophy, Leviathan (Hobbes 1980 [1651]). Religious controversies about esoteric issues had already resulted in bloody civil wars. If a few gentlemen could claim to have artificially produced something non-existent and even unthinkable, a vacuum, in their private rooms, soon new quarrels would start and before long civil wars would break out again.

Therefore Hobbes took the trouble to try to refute Boyle's claim. Based on his plenist's principles, he argued that when Boyle had exhausted the air from the glass globe, something else, a substance he called aether that surrounds the Earth and which he supposed to be so subtle that it could penetrate anything, including glass, must have filled the void. To reply to this objection, Boyle made his experiment more sophisticated (NBM: 22; Shapin and Schaffer 1985: 181ff). He put a little feather in the glass globe. If the receiver, exhausted from common air, would be filled with aether, jets of aether caused by the Earth (and thus the receiver) moving through the aether, would certainly move the feather. But the feather didn't move. All witnessing gentlemen agreed on that. So, Boyle concluded, his experiment proved Hobbes's speculative reasoning to be wrong.

So, Latour concludes, it was not an old-boys’ network of powerful friends, but a hybrid alliance of an airpump, a few gentlemen, and a little feather that made Boyle's case. Of course, this event alone cannot explain the establishment of experimental science as a source of authenticated knowledge, separate from politics, religion and morals. But ‘explanation’ is not what Latour is after. He redescribes the crucial event in the Boyle–Hobbes debate. In Latour's description, Boyle's innovative step was to apply the old repertoire of penal law and biblical exegesis to a new point. Earlier, witnesses had always been human or divine. In Boyle's texts, a new kind of witness is recognized, “inert bodies, incapable of will and bias but capable of showing, signing, writing and scribbling on laboratory instruments before trustworthy witnesses” (NBM: 23). Boyle introduced “the testimony of nonhumans” – in the above experiment, the testimony of a little feather – a kind of witnessing even more trustworthy than humans. But in the court of science, this testimony will lose its power when other considerations – e.g. political, religious or moral ones – get the upper hand. So, to allow nonhumans to provide testimony, all connections with political and religious issues had to be broken off. To allow experimental science to produce authenticated knowledge about matters of fact, a two-houses solution was invented: in one house, matters of natural philosophy, that is, science, would be decided; in the other house, all other issues. “Not meddling with Divinity, Metaphysics, Moralls, Politicks, Grammar, Rhetorick, or Logick” in the house of science was an offer the experimental scientists were more than willing to make.

The dice had been cast. Powers, competences and responsibilities were distributed. A divide had been instituted. By introducing the testimony of nonhumans, Boyle and his followers had found a way to scientifically represent facts of nature – objects. In Leviathan, by creating the citizen who in exchange for safety and peace cedes rights to the artificial construction of the Sovereign, Hobbes had invented the modern political and social subject. Simultaneously, a differentiation of (scientific) content and (social and political) context had been created. This, Latour claims, is what Shapin and Schaffer's Leviathan and the Air-Pump documents – provided that their book is read as a study of the simultaneous birth of both modern science and modern politics, each with their own vocabularies, procedures, institutions and vocations. Read in the right way – that is to say: not as an exercise in sociology of scientific knowledge but as a comparative, anthropological one – Leviathan and the Air-Pump shows that the divide between science and politics was created out of the translation work performed by Boyle (on the object-side) and Hobbes (on the subject-side). Once the divide was in place, “the word ‘representation’ [would] take on two different meanings, according to whether elected [human] agents or [nonhuman] things are at stake” (NBM: 29). From now on, human values and interests were to be represented in politics, and facts of nature by science. Each had got its own institutions, procedures, rituals and meeting rooms.

5.2 The modern Constitution

How did the distribution of competences and responsibilities that had been reached in a small circle in the seventeenth century spread out to become virtually unquestioned in the centuries to come? That part of the history of the West has still to be written. To provide an answer to the question, Latour shifts registers. Leaving ethnography for the moment, he engages in philosophical analysis and speculation. He presents the arrangement that came out of the debate between Boyle and Hobbes as being laid down in a ‘modern Constitution’ that separates the powers of politics and science, to argue that the arrangement remained in place for centuries because the internal, intellectual structure of this Constitution shields it from criticism and change.

Political constitutions divide the various branches of government. They are the outcome of long political discussions before being ordained and established. Typically, they contain procedural guarantees to prevent the casual or frivolous overthrow of the established constitutional arrangements, for example by demanding that amendments to the constitution require new elections and a two-thirds majority in both chambers of parliament. The Constitution Latour writes about (Constitution-capital-c to distinguish this one from political constitutions) does not separate branches of government, but the powers of, on the one hand, ‘Nature’ – the common denominator for objects and of brute natural facts, the domain of the natural sciences – and, on the other hand, ‘Society’ – comprising human interpretation and activity, values and symbolic constructions, the domain of politics and culture. This Constitution didn't get shaped in political discussions. It evolved out of the mainstream of seventeenth, eighteenth and nineteenth-century philosophy and was anchored in the division between the institutions of science and of politics.

What Latour calls ‘the modern Constitution’ is an aggregate that introduces a host of – partly – overlapping dichotomies that contrast mind and matter; humans and nonhumans; values and facts; the Subject of knowledge and the known Object; human society, culture and politics on the one hand, and nature and science on the other hand. One would search in vain for a single philosopher who defends all of these dichotomies together in raw form; Latour's ‘modern Constitution’ is a blend of Descartes’ ontological distinction between the realm of thinking (res cogitans) and the realm of things (res extensa), and Kant's critical philosophy that argued that the world we know is the phenomenal world, the world of our experiences mediated by the pure forms of intuition and the categories of human understanding, and not the (noumenal) world of things-in-themselves. Rather than taking the whole of the modern Constitution on board, after Kant philosophers took it as the starting point for their reflections (cf. e.g. Schnädelbach 1983). With Kant, they took for granted that Newton had shown that science can inform us about the natural world, to subsequently shift their attention to the Subject-side of the divide, to mind, culture, society and language, to the foundations of the Geisteswissenschaften and to the question of what remained to be said about religion and ethics once reason had entered “the secure path of a science” (Kant 1956 [1787]: Bix). Their attempts to bridge the gap between Sein und Denken, between the Object- and the Subject-pole, started from the Subject-pole. Idealism reigned. Even philosophers who usually are not labelled as ‘idealists’ (e.g. the early Wittgenstein, cf. Stenius 1964) followed the strategy of deducing whatever they had to say about the ontological structure of the world from their reflections on knowledge, language and mind.

True, not all philosophers were convinced and the nineteenth-century discovery that not only do societies and cultures evolve but that also geological formations and animal species had changed over time presented quite a puzzle (Mandelbaum 1971). ‘Life’ became a heated subject of debate in nineteenth-century philosophy; vitalists argued that living creatures have their own ontology, separated from inanimate nature. Schopenhauer (1977 [1818]) boldly stated that to deny nature to be nothing more than appearance would be a form of solipsism that belongs only to the madhouse. In his philosophy, not only humans but also inanimate objects show a ‘Will’. Dissociating ‘will’ from its psychological connotations, straining language, Nietzsche even argued that inanimate beings are (not: have) a ‘will’ of their own (cf. Nehamas 1985: ch. 3). But the mainstream of the European tradition placed the Object-side, Nature, beyond philosophical analysis. To be informed about Nature, one should consult a scientist, not a philosopher. In spite of generations of philosophers arguing for more subtle distinctions, the modern Constitution – the fundamental divide between the poles of Nature and of Society – was firmly in place.

Of course, the philosophers didn't deny that there is a world out-there to which humans intimately connect and that shapes human relations: people need food, water, energy, they share a house, operate machines together, exchange material items and use a wide range of technologies to communicate and connect with each other. If there were not a material world, society would not last a minute and without their bodies being fed, human minds would soon have other concerns than theoretical philosophy.

So why did the modern Constitution remain firmly in place, in spite of generations of philosophers arguing for more subtle conceptual distinctions and in spite of the mundane considerations that speak to the contrary? On Latour's speculative analysis, like a political constitution, the modern Constitution, too, came with guarantees that prevented overthrowing its arrangements light-heartedly. The internal architecture of the modern Constitution led us into an intellectual game the outcome of which is decided: ‘Heads I win, tails you lose.’

These are the rules of this game, the “guarantees of the modern Constitution” (NBM 29–35). Guarantee (1): The Constitution declares that (ontologically) Nature is not our construction (i.e. Nature is transcendent), but Society is (i.e. Society is immanent). Nevertheless, with concern to knowledge, the tables are turned. Guarantee (2): The Nature science talks about is the phenomenal world, mediated by the categories of human understanding and often artificially constructed in a laboratory (Nature is immanent); whereas Society is given as an ensemble of social facts that social scientists may investigate (Society is transcendent). So the Constitution paradoxically declares: even though we construct Nature, Nature is as if we did not construct it. A similar paradox applies at the other side: though we do construct Society, Society is as if we do not construct it. Both paradoxes are evaded by Guarantee (3): There shall exist a complete separation between the natural world and the social world; and there shall exist a total separation between the work of translation (that goes on in constructing nature in laboratories and society in social practices) and the work of purification (that goes on in stating that the two spheres are ontologically separate). Finally, Latour lists Guarantee (4): God is removed forever from the dual social and natural construction. But, Latour observes, this guarantee is offered only half-heartedly. God remains presentable and usable. In case of conflict between the laws of Nature and those of Society, the right to appeal to the transcendence of God is reserved. As an effect, spirituality was reinvented: “[m]odern men and women could […] be atheists while remaining religious” (NBM: 33) – although perhaps only in their heart of hearts.

With these four guarantees, the modern Constitution presents simultaneously a set of anchor points and a set of possibilities that provide an unbeatable set of four contradictory resources for critique (NBM: 36). Nature's laws are given, but we have unlimited possibilities to experiment and to develop technologies; men are free, but social science may establish the laws of human behaviour and stable patterns of social action we can do nothing against. So Moderns “can mobilize Nature at the heart of social relationships, even as they leave Nature infinitely remote from human beings; they are free to make and unmake society, even as they render its [social, economic and psychological] laws ineluctable, necessary and absolute” (NBM: 37). ‘Heads I win, tails you lose.’

However, in spite of their dualistic worldview, the Moderns continued to mix humans and nonhumans “under the table”, in their translation work. They created an abundance of new hybrids. Laboratories kept producing knowledge that could be used for developing technologies that increasingly began to make up the fabric of modern society – vaccines, technologies for long distance communication, plastics, computers, genetically modified organisms, etc. However, in the ontology of the modern Constitution there is no place for these hybrids. So they were purified. The results of the translation work in scientific laboratories were taken for ‘discoveries’ of brute facts of Nature, attained by scientists with rational minds and methods. Also the technologies that came out of laboratories were purified. If their origin was discussed, they were conceived as the products of value-free science; but their use and social impact were conceived as a matter that might be discussed in politics and studied by social scientists. After this clean up, no further discussion was deemed necessary. Amazingly, the role of technology in making up modern societies and in establishing social order was not studied. As to philosophy of technology: it never took off.

The Constitution's Guarantee (3) had made the practice of translation invisible, under the radar of the official view. But this didn't limit the practice of translation in any way. It continued at great pace. Without it,

[…] the modern world would immediately cease to function. Like all other collectives [the term Latour uses to designate associations of humans and nonhumans] it lives on that blending. On the contrary (and here the beauty of the mechanism comes to light), the modern Constitution allows the expanded proliferation of the hybrids whose existence, whose very possibility, it denies.

(NBM: 34)

The Constitution had another important effect. Apart from the divide between Nature and Society, it also created a divide between Us, Moderns, and Them, premodern peoples, who continue to confuse the domains of humans (Society) and nonhumans (Nature) that the Constitution declared to be separated. Placing events either before or after the advent of the Constitution, the Constitution created its own time-line with a clear break. Before: traditionalism, confusing things and men; after: modernity and the future, separating the disenchanted world of brute facts from the domain of the mind, values and meanings. The internal divide between Nature and Society seamlessly led to an external divide between Us (Moderns) and Them (premoderns).

Weber's analysis of the exceptionality of modern Occidental culture and society is a straight consequence of identifying modernity in the terms of its Constitution. Weber was right, that is to say: once we accept the terms of the modern Constitution. No, Weber was wrong: we have never been modern; next to the practice of purification the practice of translation has also continued. And this practice led to new concerns, partly reformulating the two downsides of modernity Weber had already observed – loss of freedom and loss of meaning – but also introducing new ones: the scientification of politics, the ideology of industrial society, efficiency, and ‘one dimensional man’ (Marcuse 1964). What Weber had singled out as the Occident's exceptional form of rationality, Zweckrationalität, came to be perceived as a ‘halved rationality’ (Habermas 1969). Where could a broader concept of rationality be found? In search of it, the Frankfurt School and other critical theorists again focused exclusively on the Subject-pole, fortifying the modern Constitution further.

Latour displays little patience with critical theory. He takes a more radical turn. He sets out to introduce a conceptual framework that will let us see what the Moderns, looking through the lens of the modern Constitution, failed to see. To the practice of purification, he adds the practice of translation, the practice from which hybrid beings emerge, to give it pride of place. The ontology developed in almost twenty years of science studies was to replace the modern Constitution. Kant had revolutionized philosophy by turning Descartes’ philosophy inside-out; Latour's philosophy drains the modern Constitution by prioritizing the middle ground, the production of hybrids, rather than the two poles of Nature and Society. He replaced the modern Constitution by what has been called “hybrid thoughts in a hybrid world” (Blok and Jensen 2011).

5.3 Relationism

Hybrid thoughts? For a philosopher, that is not necessarily a compliment. So let's first describe in broad strokes what Latour suggests as an alternative to the modern Constitution, to subsequently get to the details.

What are the ‘hybrids’ about which Latour talks? The term itself shows embarrassment. A ‘hybrid’ denotes something made by conceiving two different elements, a mixture – but a mixture of what? Humans and nonhumans? Nature and Society? Construction and reality? Subject and Object? But then we define ‘hybrids’ in terms of the language of the modern Constitution – the conceptual architecture that, Latour suggests, needs to be replaced by something else.

Again, Latour runs against the problem anyone encounters who wants to formulate a radical new idea: he has to communicate his new thoughts in the old language. In his work in science and technology studies, Latour could take his readers by the hand. The ethnographic material he brought to the fore allowed him to point his readers to what they had failed to see before; meanwhile he could introduce, step-by-step, a new terminology for describing what he wanted his readers to notice. In We Have Never Been Modern – and in his later work An Inquiry into Modes of Existence (cf. Chapter 6) – Latour presents the conclusions of his earlier work in science and technology studies in general terms. He wants his modern readers to get rid of the conceptual framework that comes naturally to them and that is officially institutionalized in the separation of the representation of humans by politics and of nonhumans by science; he wants them to abandon the conceptual distinctions that evolved out of the attempts of eighteenth and nineteenth-century philosophers to account for science in epistemological terms, that is, in terms of the relation between an active knowing Subject and inert Objects lying waiting for centuries to be ‘discovered’ by a scientific genius. He asks them to replace the Weltanschaung that comes naturally to them with a philosophy that prioritizes existence over essence (i.e. eternally given – but hidden – substance) and that gives ontology pride of place over epistemology. He wants them to turn their Weltanschaung inside-out, to define hybrids not in terms of the constituents the modern Constitution takes for being given, but rather to give them pride of place and to see what then appears as given, as the outcome of processes, of translations.

But how to argue for this transition without using the old terms? Latour asks his readers to acknowledge that we are uncertain about what is action, a social group, a fact, technology, what is law, science and politics and to abandon not only what they thought they knew about knowledge, but also what they thought they knew about the world. To make this move, Latour invites his readers to have a closer look at who we are and what we practice.

As we have seen above, the Constitution's internal divide between Nature and Society led to an external divide between Them – premodern peoples – and Us – the Moderns. The Moderns locate themselves on a timeline that runs from past to present – and further into the future – with a radical break in the past, the birth of modernity; a break that took place somewhere in the sixteenth and seventeenth centuries, when – as Weber suggested – Platonism married Renaissance experimentalism to change Man's conception of the universe and his own place in it, while simultaneously modern politics was invented by Hobbes.

However, as Latour points out, our actual practices are more heterogeneous.

I may use an electric drill, but I also use a hammer. The former is thirty-five years old, the latter hundreds of thousands. Will you see me as a DIY expert ‘of contrasts’ because I mix up gestures from different times? Would I be an ethnographic curiosity? On the contrary: show me an activity that is homogeneous from the point of view of the modern time. Some of my genes are 500 million years old, others 3 million, others 100.000 years, and my habits range in age from a few days to several thousand years. As Péguy's Clio said, and as Michel Serres repeats, ‘we are exchangers and brewers of time’ (Serres and Latour 1995). It is this exchange that defines us, not the calendar or the flow that the [M]oderns had constructed for us.

(NBM: 75)

Brewers of time? So are we still confused, living for one bit in modernity and for another bit in traditionalism? No. We need a more refined conception of time. The whole idea of a stable ‘tradition’ is a misconception. Most ‘traditions’ are modern inventions; for example, the traditional Scottish kilt was invented at the beginning of the nineteenth century (Trevor-Roper 1983). “One is not born traditional; one chooses to become traditional by constant innovation” (NBM: 76).

Nevertheless we actively sort out elements belonging to different times. But to sort events in time, one has to first introduce calendars and clocks. “It is the sorting that makes the times, not the times that make the sorting” (NBM: 76). Calendar time orders events on a chosen timeline. Crucial events are given a date. 1812: Battle of Borodino. 1858: Pasteur discovers ‘lactic yeast’, the birth of microbiology. 1995: the first United Nations Climate Change Conference. We may also put dates to the emergence of new geological formations or to the evolution of a species on our calendar, e.g. to record that the geological epoch of the Holocene started 11,700 years before the present.

But calendar time is not ‘history’. As we have seen before (ch. 3), for Tolstoy ‘history’ does not denote a series of events in calendar time, but infringements of life. Latour cites a similar idea from Péguy's Clio. Comparing Victor Hugo's “terrible” play Les Burgraves with a little phrase from Beaumarchais, Péguy (cited in NBM: 68) wrote:

When I am told that Hatto, the son of Magnus, the Marquis of Verona, the Burgrave of Nollig, is the father of Gorlois, son of Hatto (bastard), Burgrave of Sareck, I learn nothing,’ she [Clio, the muse of history] says. ‘I do not know them. I shall never know them. But when I am told that Cherubino is dead, in a swift storming of a fort to which he had not been assigned, oh, then I really learn something. And I know quite well what I am being told. A secret trembling alerts me to the fact that I have heard.’

A ‘historical’ event is an event that happens to someone, an event that affects one's existence. ‘History’ situates events with respect to their intensity. For Tolstoy ‘history’ is limited to the human domain. It denotes an event that happened to someone's, a group's, or a nation's existence. Latour suggests that the notion of history also applies to nonhumans, to note that something has happened to it – that is, has affected its existence. And he claims that it not only applies to nonhumans that – as we know since the nineteenth century – have evolved over time (like biological species and geological formations), but to any nonhuman and (when we put calendar dates to the events that happen to them) on scales that are much smaller than geological or evolutionary time. For example, when a nonhuman is put on trial in an experiment, this event may become part of the history of that nonhuman.

To get the idea, reconsider the discovery of lactic yeast by Pasteur. What is the object Pasteur discovered in his experiments? As we have seen before (§§ 2.3 and 2.4), it started as a barely visible grey mass in ordinary lactic fermentation, was turned into something Pasteur could collect, transport and sprinkle, was subsequently identified by Pasteur as a living organism – like brewer's yeast but different – to be turned in experiments of later generations of microbiologists into ‘lactic acid bacteria’ (plural), that is, not one specific micro-organism but a whole clade of bacteria.

The modern Constitution suggests that in the course of a century of research humans have discovered the true nature of the microbes that cause souring of milk, their essence. By now, we know that what Pasteur called ‘lactic yeast’ is, in fact, a whole clade of microbes, rather than a specific member. That, however, is the epistemological interpretation – that is, the story of the way in which in the course of (calendar) time the Object appeared to the knowing Subject.

Latour suggests that to account for what has happened we better tell this story in ontological terms. In that story, events happened that changed this being's existence. It had quite some career. It started as a ferment of ordinary lactic fermentation (a dubious – chemical or biological? – entity that even might not exist at all); it became the barely visible grey mass that drew Pasteur's attention and that eventually led him to claim it to be a kind of yeast; to become a hundred years later a family of bacteria. What happened to this being? In 1857, it met Pasteur, who provided a culture in which it could multiply at greater speed then ever before. As a consequence, it acquired a new characteristic: it became a clearly visible being. Some decades later, in the experiments of microbiologists of a later generation, it changed its existence again to become a whole family of bacteria. That is the history of this being.

Sartre (1966 [1945]: 32–33) famously wrote “l'existence précède l'essence”. Man first of all exists, encounters himself, surges up in the world – and defines himself afterwards. Latour extends the phrase to apply also to nonhumans: their existence precedes their essence too. In the course of its history, the existence of a nonhuman may become more stable, at which point we may guess that we have identified its essence (while acknowledging that future research may correct or even overthrow this guess – yes, we have read Popper). But in ontological terms, nonhumans also have a history; they are affected by what happens to them. They are not the Object of the modern Constitution, an essence waiting to be discovered; they are beings whose existence changes in the course of their history. They have a ‘variable ontology’ – their characteristics as an existent change when something happens to them.

Of course, we may subsequently put some dates next to the events that happened to a being, to locate these events in calendar time. So we may speak of the being that in 1858 was identified as ‘lactic yeast’ and about the being that more than a century later is identified as ‘lactic acid bacteria’. However, to carelessly mix up calendar time with history, thinking that although time has flown it has remained the same being with a stable substance (that is, the Object the modern Constitution), is asking for conceptual trouble. No, Pasteur's lactic yeast and today's lactic acid bacteria have different ontologies. In ontological terms, they have quite different existences; they are different beings.

Moving back in time, this leads to a conclusion that sounds completely weird to modern ears: ‘lactic yeast’ did not exist before Pasteur. Of course, before Pasteur, milk, when left alone for a few days, went sour. But to say that in the old days either ‘lactic yeast’ or ‘lactic acid bacteria’ caused the souring of milk is an anachronism. Whatever caused milk to turn sour before Pasteur became ‘lactic yeast’ – a being with new characteristics, namely a being that is visible to the human eye and that can be isolated, sprinkled and transported – only after having been translated in Pasteur's experiments (PH: ch. 5; Latour 1996c; 1999c; 2008).

The concept of ‘quasi-object’ bends the language of the modern Constitution for the purpose of understanding the ontology Latour is after. It may help understanding what Latour is after, but may also cause confusion. Isn't there a better term? In the 1990s, Latour suggests ‘proposition’, a concept, drawn from Whitehead, that does not designate a statement that may be true or false, but any actant (human, nonhuman, or any hybrid association of humans and nonhumans) that proposes itself to other actants to become translated into another (joint) being (PN: 247; PH: 309; the definitions differ slightly). But this term, too, didn't last long. It took Latour twenty-five years before he found in An Inquiry into Modes of Existence the words to express the difference: they are beings-as-other, rather than (the modern Constitution's) beings-as-being. Terrible jargon, indeed. We'll get to it in §6.1.

Having introduced ‘quasi-objects’ to replace the Object of the modern Constitution, we may expect Latour also to replace the Subject of the modern Constitution by ‘quasi-subjects’. In this case, we may seem to have less trouble in understanding what is meant. Nineteenth-century philosophy already had argued that the Subject has history; novels and Freud's psychology taught us to conceive persons in terms of their biographical past, in terms of what has happened to them as a child and in their personal relations. However, Latour's reasons for speaking about ‘quasi-subjects’ diverges from these traditions. The main reason is not to allow societies, cultures and human minds to have a history, but to allow nonhumans to be part of the existence of (quasi-) subjects. In Reassembling the Social, we have already become acquainted with the idea that actors, human beings, persons, are not stand-alone entities. They too have variable ontologies. For their existence, to act, they require plug-ins, scripts, and attachments. Their action may be overtaken; their instruments may make them do something. To speak about the existence of societies, cultures and the human mind, we have to take also the nonhumans into account. They happen also to human beings. Yes, Pasteur happened to the microbes, but the microbes also happened to Pasteur, to make him the person France would celebrate as its genius. We cannot understand who humans are, without taking the nonhumans, the quasi-objects, into account.

In a sense, by distinguishing between time and history, Latour has transferred to ontology a key idea from Einstein's theory of relativity. Time is not a given framework, Einstein taught physicists; to sort events in time, one needs calendars and clocks. For Latour, existence is relative; to register how (human and nonhuman) beings exist, one has to focus on their relations, mediated by instruments, experiments, by other humans and nonhumans, whatever, to detect what has happened to them, in their history.

So is Latour a ‘relativist’ – a reproach he has had to defend himself against time and again (PH: 4–23)? To avoid the confusions the term ‘relativism’ invokes in philosophy (where it is mainly associated with moral or epistemological relativism, poor man's philosophies that Latour never held), we better avoid this label. Better to call his ontological position relationism (NBM: 114). So to call Latour a philosopher of ‘hybrid thoughts’ is a bit unfair. He defends a systematic, coherent philosophical position, namely a relationist ontology, a philosophy that gives pride of place to relations, to translations, rather than to what enters into these relations.

It has been quite a philosophical tour de force indeed. But remember that the reasons for engaging in it were concrete questions. For a long time, we could live with the modern Constitution to benefit from its fruits. It allowed the proliferation of ever more hybrids, many of which helped to advance health and prosperity. But by approaching the end of the twentieth century, concerns about technologies, ecological problems, and advances in biotechnology and medical science that elicit ethical questions have begun to dominate the political agenda. They put pressure on the established way of understanding the relations between nature, science and politics. They signal that we need a better way of understanding what we have been doing, who we are and how to live in a world that includes both humans and nonhumans.

This is what urged Latour to dive deep into philosophical waters. “We want the meticulous sorting of quasi-objects to become possible – no longer unofficially and under the table, but officially and in broad daylight” (NBM: 143). Indeed, in this desire to bring to light, to make public, Latour adheres to the intuitions of the Enlightenment. But to get quasi-objects in public view, we have to abandon the modern Constitution that emerged out of the Enlightenment. We need a philosophy that allows detecting variable ontologies: a relationist, empirical philosophy.

5.4 Cosmopolitics

In We Have Never Been Modern, Latour advances the philosophy that he deployed – for a long time implicitly – in his work in science and technology studies to understand the political issues that the world faces today, in particular in relation to ecological problems.

New political thinking is called for. On that Beck and Latour agree. The political philosophy that came out of the modern tradition did not anticipate that governments would have to concern themselves with administrating the global climate, the ozone layer and environmental problems that transgress national borders. Conceiving politics to be exclusively the realm of human values, interests and ambitions, it focused on analysing the power-relations between citizens and the state. But ‘green’ political philosophies that have inspired ecological movements do not fare better, Latour argues. Calling on Society to protect Nature, they only seek “to position themselves on the political chessboard without redrawing its squares, without redefining the rules of the game, without redesigning the pawns” (PN: 5). They appeal to Nature as an authority, only to find scientists fiercely debating what Nature tells us. To develop a more realistic political ecology, the concept of Nature has to go; we have to reconsider what “nature, science, and politics have to do with one another” (PN: 6).

In the final sections of We Have Never Been Modern, Latour already outlined his political philosophy. It ends with what he calls a ‘Parliament of Things’ in which both humans and nonhumans are properly represented – a rather troubling concept: how are the speechless nonhumans to be represented; and if we allow scientists to do so, does Latour propose that scientists become members of parliament, like the military taking up guaranteed seats in parliament in some dictatorially governed countries? Our confusion will hardly be eased by Latour's statement that “we do not have to create this Parliament out of whole cloth […], [w]e [only] have to ratify what we have always done […]” (NBM: 144).

Fortunately, in Politics of Nature, he offers more detail. He defends a ‘political ecology’ that “has nothing at all to do with ‘nature’ – that blend of Greek politics, French Cartesianism and American parks” (PN: 4–5) – a political philosophy that designates “the right to compose a common world [of humans and nonhumans], the kind of world the Greeks called a cosmos” (PN: 8), a philosophy for cosmopolitics.

Politics of Nature is dedicated to among others Isabelle Stengers, who introduced the concept of ‘cosmopolitics’ (Stengers 1996–1997). The term is rather grandiose and needs some unpacking.

In the first place, ‘cosmopolitics’ is not to be taken for ‘cosmopolitanism’, the idea – defended in various versions by, among others, the Stoics and Kant – that all human beings are (or can and should be) citizens of a single community. Secondly, in contrast to what the term ‘cosmos’ might suggest, we do not have to look at the starry heavens and to think about the universe as a whole. We should not look up, but down. The ‘cosmos’ Latour's philosophy addresses is an assembly of mundane assemblies; it refers to the one Earth that humans share with other humans and with nonhuman entities; the concept acknowledges – with Pasteur – that there are more of us than we thought and that there is a multiplicity of interests that have to be taken into political account – human ones as well as the interests of nonhumans.

This leads to the second component of the term ‘cosmopolitics’, ‘politics’. Given the multiplicity of beings involved, to institute some order and peace, recourse to an encompassing entity or idea – e.g. Reason, Nature, or God – or some arbiter agreed-upon by all is not available. That, according to Latour, is what ‘politics’ is about: instituting order in the absence of any a priori point of agreement. With some reluctance, Latour cites Schmitt (1996 [1932]), a “toxic and nevertheless indispensable” (FG: 295) author, who claimed that “[t]he specific political distinction to which political actions and motives can be reduced is that between friend and enemy” (Schmitt 1996 [1932]: 26). Schmitt's further explanation may not necessarily comfort us:

The political enemy need not be morally evil or aesthetically ugly; he need not appear as an economic competitor, and it may even be advantageous to engage with him in business transactions. But he is, nevertheless, the other, the stranger; and it suffices for his nature that he is, in a specifically intense way, existentially something different and alien, so that in the extreme case conflicts with him are possible. These can be neither decided by a previously determined general norm nor by the judgment of a disinterested and therefore neutral third party.

(Schmitt 1996 [1932]: 27)

In spite of Schmitt's bellicose language, this sets the stage for what is the lead-question of Latour's cosmopolitics: how to establish a common, yet plural world, one in which we are confronted and challenged by – human and nonhuman – others, without taking recourse to some pre-given, agreed-upon authority?

For the sake of clarity, the argument of Politics of Nature may be divided into four steps.

In the first step, Latour asks us, once again, to give ontology rather than epistemology pride of place. In Politics of Nature, he does so by urging us “to get out of Plato's cave” (PN: 10ff). Why does he step so far back in time? He tries to kill two birds with one stone. Plato's allegory of the cave suggested a division between on the one hand the Ideas of the True, the Good and the Beautiful and on the other the world of the hoi polloi, the common people, who are chained for life in a cave and who, taking appearance for reality, chatter about the shadows that are thrown on the wall they face. Only those few who can break away from their fetters and leave the cave will be able to see the True, the Good and the Beautiful. Bringing the news about what they have found outside back to the cave, they can start to teach and lead the hoi polloi. A powerful – and as it turned out often brutal – motive was seeded into the cultural, political and religious traditions of the West (ICON): to live a truly good life requires first and foremost the destruction of false images; it takes people who have seen the light to break the spell. Not only philosophers embraced that message. It motivated centuries of religious wars and ideological disputes.

In Politics of Nature, Latour merges Plato's philosopher with the modern scientist. That, of course, is incorrect. Modern scientists only claim to have exclusive access to nature, to Truth, not to the Good and the Beautiful. In terms of Plato's allegory, the scientific revolution and the philosophical tradition that accepted the role of the new science projected the Good and the Beautiful back into the cave, to become a matter of value-disputes, of conflicting subjective opinions.

However, two important assumptions were retained: the idea that public life is organized into two houses – which under the modern Constitution would become ‘Nature’ and ‘Society’ – and the idea that only those who have the right competencies can travel between the two. Latour argues we have to get rid of both. We have to acknowledge that there exists only one Collective, an association – or rather a gathering of associations – of both humans and nonhumans. And we have to acknowledge that apart from scientists, others may contribute to including nonhumans in the progressive composition of the Collective too. Hence, the tasks ahead are firstly to envision a Collective and secondly to describe an explicit procedure for publicly representing “associations of humans and nonhumans, in order to decide what collects them and what unifies them in one future common world” (PN: 41).

When we unpack the term ‘the Collective’, the first task turns out to be less grandiose than it may seem at first sight. The term is introduced to stress the work of collecting into a whole, that is, the work to build a common world. Immediately, Latour makes clear that he is after something quite mundane; so, we had better stop capitalizing the word ‘collective’.

The word [‘collective’] should remind us of sewage systems where networks of small, medium, and large ‘collectors’ make it possible to evacuate waste water as well as to absorb the rain on a large city. This metaphor of the cloaca maxima [ancient Rome's sewage system] suits our needs perfectly, along with all the paraphernalia of adduction, sizing, purifying stations, observation points, and manholes necessary to its upkeep. The more we associate materialities, institutions, technologies, skills, procedures, and slowdowns with the word ‘collective’, the better its use will be: the hard labour necessary for the progressive and public composition of the future unity will be all the more visible.

(PN: 59)

Remember that Latour does not want to introduce a new institution, some Parliament with a brass nameplate next to its door. He set out to redescribe what we have been doing all along but failed to acknowledge and to ratify – the practices of composing a common world. Latour is and remains an empirical philosopher. The conceptual apparatus he introduces serves the purpose of providing a language to better account for what we do and what is troubling us.

What is troubling a – or ‘the’ – collective? Why would a collective need a ‘cosmopolitics’ at all? This brings us to the second step of Latour's argument. At any moment, something can happen to a collective. Time and again, new entities knock on its door, offering a proposition (in Whitehead's sense of the term), asking (if not pressing) to become included. The appellant may be a virus, a new technology, immigrants who have been risking their lives to cross the Mediterranean or the US border, a species that signals that it is threatened to become extinct, a so far unidentified agent that causes a cattle-disease – anything. Whatever it is, the collective sees itself confronted with two questions. In the first place: how to take this appellant's proposition properly into account? Secondly, can we live together? The first question requires the proposition to be adequately articulated. The second question requires reflection on how the appellant would fit into the collective. Finally, the question has to be answered whether the appellant should be internalized as a member of the collective or externalized, that is, rejected.

To give appellants fair and proper treatment, to guarantee “due process”, Latour formulates four requirements (PN: 102–109). The first two pertain to the first question; the third and fourth to the second one. They are addressed to the collective that sees itself confronted time and again with new propositions.

(1) You shall not simplify the number of propositions to be taken into account in the discussion. Latour calls this first requirement the requirement of perplexity.
(2) You shall make sure that the number of voices that participate in the articulation of propositions is not arbitrarily short-circuited. This is the requirement of consultation.
(3) You shall discuss the compatibility of new propositions with those that are already instituted, in such a way as to maintain them all in the same common world that will give them their legitimate place. This is the requirement of hierarchization.
(4) Once the propositions have been instituted, you shall no longer question their legitimate presence at the heart of collective life. This is the requirement of institution.

We may note that the first and fourth requirements – roughly – cover what was formerly contained in the notion of ‘fact’; and that the second and third requirements cover what was formerly contained in the notion of ‘value’. Latour grants that the Moderns have correctly intuited that questions about is and ought need to be separated, but he thinks that this intuition is inadequately covered by being formulated in terms of facts (the domain of science) and values (the domain of politics). He replaces the distinction between is and ought by an alternative one (PN: 102). Where in the modern Constitution facts and values were dealt with respectively in the house of Nature (science) and the house of Society (politics, culture), he suggests that due process requires a new bicameralism, a clear division between on the one hand “the house for taking into account” (that will deal with perplexity and consultation) and “the house for arranging in rank order” (that will deal with hierarchization and institution) (PN: 153). Schematically represented, the separation of powers the modern Constitution suggested is rotated by ninety degrees (PN: 115).

Provided the four requirements are fulfilled and the separation between, on the one hand, the house of taking into account, and, on the other hand, the house of arranging into order is respected, when appellants propose themselves, they enter a procedure that will give their propositions due process. Their propositions will be taken into account (perplexity and consultation) and ranked (hierarchization and institution). If instituted in the collective, they will become a legitimate member of the collective; if not they will be externalized, rejected, but not forever. They may appeal again to be included; cosmopolitics is a continuous process, the path from perplexity, consultation and hierarchization to institution will have to be iterated time and again.

Apart from the four tasks already mentioned – (1) perplexity, (2) consultation, (3) hierarchization and (4) institution – cosmopolitics requires two other tasks, namely (5) to keep a clear separation between what goes on in the ‘house of taking into account’ and the ‘house of arranging into order’ and (6) to provide the collective with some self-image, a “scenerization of the totality” (PN 137–8), that defines (for the time being) what is inside and outside the collective.

Whatever ‘scenerization of the totality’ the collective will design, it will be clearly different from the image provided by the modern Constitution. The image of a Society surrounded by Nature is replaced by a distinction between what is internalized (i.e. instituted in the collective) and what has been externalized. The exteriority is no longer fixed and inert, as Nature was to Society under the modern Constitution. If an appellant is externalized from the collective, it may (and probably will) propose itself again in a new iteration of the process. So the exteriority of the collective is no longer a reserve (‘Nature’), nor a court of appeal (‘hard facts’), nor a dumping ground (‘the environment’) for Society. The outside of the collective consists of entities that may appeal to be included, and that may either threaten or enrich the collective. Latour's collectives are explicitly open to contingencies, innovation and change.

By formulating this procedure for the progressive composition of a common world of both humans and nonhumans, Latour does not propose to overthrow existing practices, to formulate an utopia or call for revolution. His ambition is to make explicit what we have been doing all the time, but implicitly, under the table.

To see this, consider, once again, the role of Pasteur in society. In the mid-nineteenth century, the variation in virulence of contagious diseases presented quite a problem. That was the problem that perplexed hygienists and many others and that got Pasteur started. He succeeded in identifying microbes as the cause of diseases, studied their characteristics and determined how they contributed to spreading diseases among other members of the collective – humans and cattle – (consultation), subsequently to suggest ideas about how to contain microbes so as to prevent them from communicating diseases (hierarchization). Technical facilities had to be designed, rights and burdens to be redistributed, new hygienic forms of behaviour to be facilitated and stimulated, and so on. Once that had been implemented – in cities, households, farms and in industry – the collective had both instituted the undeniable existence of microbes and had learned how to live better with them.

Latour's cosmopolitics turns Pasteur's practice into principle. If we focus on what Pasteur did, on his work, on the different tasks he took up, we notice that the idea that his role as a scientist was to produce knowledge that others, politicians, may use to attain their ends, only gives a very limited view on what he actually achieved. Pasteur was active in all four quadrants of the scheme that summarizes the cosmopolitical process. But he was certainly not the only one who did so. Politicians, hygienists, medical professionals and many others also contributed to each of these four quadrants. But each of them did so in their own way. To differentiate their roles, the fact/value distinction is too clumsy. We should discuss the different skills and instruments scientists, politicians and many others bring to the cosmopolitical process, that is, focus on the kind of translations they are able to perform.

This brings us to the third step of Latour's argument. Consider first the role of the sciences in progressively composing a collective (PN: 137–143). As Pasteur showed, they may contribute to each of the four quadrants. They can bring to the task of (1) perplexity the asset of their curiosity and open minds, plus laboratory-instruments that will allow them to detect scarcely visible phenomena and formerly unknown entities and powers. By designing suitable experiments to articulate the characteristics and powers of these entities, scientists can also contribute to the work of (2) consultation. Moreover, for (3) hierarchization, they have the competence to offer heterogeneous innovations and compromises that may help to reduce conflicts. Finally, scientists also have the knack of knowing how to (4) institute an entity. They know that when all is said and done, outcomes of research have to be accepted as ‘hard facts’. With regard to task (5), separation of powers, Latour points to the right of anybody to ask their own questions in their own terms, whether they are perceived as reasonable and realistic or not by others, a right to which scientists traditionally appeal when they defend their autonomy. With respect to (6) scenerization, scientists can multiply the great narratives ‘from amoeba to Einstein’ and ‘from Plato to NATO’ that provide a simplified but coherent image of the common world.

Consider, secondly, the role of politicians (PN: 143–154) – the term referring here to statesmen, parliamentarians, union leaders and the like. They too contribute to each of the six tasks, but with different skills from scientists. They can contribute by being attentive to the fact that excluded entities can return to haunt the collective (perplexity); their role is to form concerned parties, reliable witnesses, and to mobilize stakeholders to ensure a wide variety of voices (consultation). They have the skill to compromise and to persuade the ones they represent to accept them (hierarchization) and the craft to make painful decisions, accepting the fact that building a collective also means to exclude and to make enemies (institution). They know that deliberation and decision-making have to be clearly distinguished (separation of powers). They may provide a clear sense and a narrative of who we are as a collective and what for the time being must remain outside (scenarization).

In Reassembling the Social even the ‘sciences of the social’ are granted a productive role in cosmopolitics. They, and in particular the ‘cameral sciences’, have innovated forms, standards, plug-ins and centres of calculation that may help in consultation, hierarchization and institution. Their ‘panoramas’ may contribute to scenarization by providing a “prophetic preview of the collective” (RAS: 189). The latter should however be viewed with scepticism: the concept of Society on which they rest obstructs the view on the process of cosmopolitics; and “[t]o put it bluntly: if there is a Society, then no [cosmo-]politics is possible” (RAS: 250).

What has been said for scientists, politicians and sociologists of the social also holds for other disciplines. Economists, for example, provide a common language allowing commensurability and calculation that may be useful in (3) hierarchization; while moralists stress the scruples that make it necessary to go looking for invisible entities and appellants (perplexity) and the need of resumption (scenerization).

Two other types of professionals will also be required, namely administrators and diplomats. “Administrators are going to have responsibility for distinguishing all the functions […] and for coordinating the various professions” (PN: 205). Diplomats – although always belonging to one party of a conflict – are needed for their competence of sensing what the interests and moving space of the other party are. Together with administrators, they perform a seventh task, namely to ensure that due process takes its course, that the collective remains open and that the issues it is confronted with are seriously followed up.

Latour does not propose a utopia, nor does he call for a revolution. ‘Cosmopolitics’ is already widely practised, but political philosophy simply failed to acknowledge its existence. Pasteur's message that there are more of us than we thought has been incorporated into political practice already for more than a century. All over the world, scientists, economists, politicians and others meet in hybrid assemblies – in Brussels, in Washington and in any other capital. In contemporary politics much more is taken into account than the power, the interests and the values of people that traditional political philosophy highlights; also the power of microbes, noxious chemicals, ecosystems, and the global climate are at stake. Can all of these powers fit into an ordered, collective, peaceful existence? As Koch and Pasteur showed: it may require quite some reordering, a combination of technical, economic, social and legal measures. Today, apart from politicians and the citizens they represent, scientists, economists, jurists and moralists also bring their skills and instruments to contribute to political deliberation and decision-making, that is, to each of the four quadrants of the (cosmo-)political process. Cosmopolitics is already with us, in and outside traditional political institutions, on municipal, national, European and global levels. We only failed to acknowledge its existence properly.

In Politics of Nature, Latour is redescribing the existing practices of composing a common world with particular focus on the role of science. He redescribes the role of science in democracy. He is not pleading for ‘democratization of science’, that is, for including the voice of non-scientists in the production of scientific knowledge, a subject widely discussed among scholars in science and technology studies. No doubt also non-scientists can and have to contribute to cosmopolitics, for example by bringing their practical skills and experience to the job and by mobilizing new voices, locally and via the internet. But this input should not be mistaken for contributing to science; to contribute to science, specific skills, instruments and access to laboratories are required, which in most cases are out of reach for non-scientists.

This brings us to the fourth and final step of Latour's argument: the empirical evidence that cosmopolitics is already practised. In Politics of Nature, apart from scattered examples, the reader has to consult the literature mentioned in footnotes to get a view of the evidence. But in 2005, Latour curated Making Things Public, an exhibition in Karlsruhe's Zentrum für Kunst und Medientechnologie (ZKM), that showed an abundance of illustrations of the practices and instruments of cosmopolitics collected by Latour and his many collaborators.

The introduction to the exhibition's hefty catalogue (MTP), shows a slight change in terminology (yes, Latour keeps innovating, fine-tuning his terminology). We Have Never Been Modern ended with a call for ‘a Parliament of Things’. Its readers may have thought that ‘things’ were (quasi-) objects, to wonder how speechless nonhumans were to have any say at all, or, alternatively, whether Latour thought that apart from elected politicians, scientists should also become Members of Parliament, to represent the speechless inanimate masses, a suggestion that would immediately lead to the question how the election of these representatives should be organized. The catalogue's introduction (MTP: 14–41) removes the misunderstanding. A ‘thing’ is not an object, a Ding an sich, nor the ‘things’ of common sense, a Gegenstand (a stone, your car, etc.). The term is used by Latour in its Old English meaning, to designate a gathering, a meeting, a court, that is, any place, where a ‘matter of concern’ is made public. In some European countries, the old meaning has remained in use. In Norway, the parliament is called the Storting, in Iceland deputies gather in the Althing. Heidegger used the same archaic meaning to distinguish between Gegenstand and Ding (Harman 2005: 268–271). So a ‘Parliament of Things’ is a pleonasm, or a funny name for what in Politics of Nature is called the aggregate of assemblies involved in cosmopolitics. It is not one place, but a whole array of distributed ways for progressively composing a common world of humans and nonhumans by making matters of concern res publica.

Drawing from scientific laboratories, technical institutions, marketplaces, voting practices, churches and temples, internet forums, ecological disputes, and from practices outside the ‘modern world’, the exhibition at ZKM showed the variety of

forums and agoras in which we speak, vote, decide, are decided upon, prove, and are being convinced. Each has its own architecture, its own technology of speech, its complex set of procedures, its definition of freedom and domination, its ways of bringing together those who are concerned – and even more important, those who are not concerned – and what concerns them, its expedient way to obtain closure and come to a decision.

(MTP: 31)

We have never been modern; we have continued the practice of translation, creating ‘hybrids’ all the time. And we also have been practising cosmopolitics for more than a century. But we have been doing both “under the table”, carelessly, without considering it as part of the challenge to progressively building a common world.

In Politics of Nature, Latour emphasizes that he has “no utopia to propose, no critical denunciation to proffer, no revolution to hope for […]. Far from designing a world to come, [he has] only made up for lost time by putting words to alliances, congregations, synergies that already exist everywhere and that only the ancient prejudices kept us from seeing” (PN: 163). He has redescribed our political practice, “asking for a tiny concession: that the question of democracy be extended to nonhumans” (PN: 223), namely by explicitly acknowledging what we already do clumsily. In cosmopolitics the nonhumans are allowed to speak up, by authorizing a wide range of representatives (scientists, politicians, economists, moralists) to speak for them, that is, to articulate their propositions and to answer the question whether and how it is possible to live with them together.

Cosmopolitics is here and is here to stay. Visit any of the major institutions that help govern the world – the FAO, the World Health Organization, the OECD, the IPCC, the World Bank and the IMF, or the public authorities that protect the environment and the safety of food – or visit any ministerial department, and you will meet scientists, economists, lawyers and politicians in conference in hybrid assemblies. The forms and constitutions of these assemblies vary; most of them are quite recently erected. And that, of course, allows us also to imagine other forms of assemblies and to take initiatives, to try to innovate and to form new ones. Once this has been acknowledged, the question can be raised how we could better institute cosmopolitics, to become more attentive to what we have been doing, to develop new competences that will be needed to progressively compose a common world. There is no grand design, nor a Providence, to guide us, nor any other authority on which we can rely. We will have to find out, empirically, experimentally.

By showing that we have never been modern and by asking us to ratify the procedures we have been following all the time, although only implicitly and often without the necessary diligence, Latour defends our civilization – the civilization that brought us, among others, science, the rule of law, modern ethics and politics – that is to say: provided ‘civilization’ is not understood in terms of modernity's ‘progress’, but defined “by the civility with which a collective allows itself to be disturbed by those whom it has nevertheless explicitly rejected” (PN: 208–209).

But this dual position towards modernity opens up a new question. If we need the competences of scientists, politicians, moralists, jurists and economists for this task, does that imply that we have to endorse the modern values of Science, Politics, Morality, Law, and Economy? And how should we account for these values, once the conventional accounts, based on the modern Constitution, are discarded? In An Inquiry into Modes of Existence, published twenty years after We Have Never Been Modern, Latour set out to address these questions. If not modern, what or who are we? How can we propose and defend the values we hold dear – the values “that the notion of modernization had at once revealed and compromised” – in the “planetary negotiation that is already under way over the future of the[se] values” (AIME: 17)?