In favor of a dialogue between political science and Science Studies

ABSTRACT

Political science and Science Studies both use the words “politics” and “science,” yet each applies such different meanings as to nearly appear incommensurable. Here, I explain to political scientists the various uses of these words by Science Studies practitioners and how they could relate to those used by political science. The paper shows that only one of these meanings of the word “science” (out of four possible meanings) distinguishes it in a radical way from “politics” (which may take on six different meanings). Once those meanings have been defined, I propose that a more fruitful collaboration between Science Studies and political science could be developed based on the method of “following issues,” rather than delimiting distinct domains for science and politics.

RESUMO

Tanto a ciência política quanto os Estudos Sociais da Ciência usam as palavras “política” e “ciência,” mas cada uma delas aplica significados tão diferentes que quase parecem incomensuráveis. Aqui, explico aos cientistas políticos os vários usos dessas palavras pelos profissionais dos Estudos Sociais da Ciência e como eles podem se relacionar com os usados pela ciência política. O artigo mostra que apenas um desses significados da palavra “ciência” (de quatro possíveis) a distingue de forma radical de “política” (que pode assumir seis significados diferentes). Uma vez definidos esses significados, proponho que uma colaboração mais frutífera entre os Estudos da Ciência e a ciência política poderia ser desenvolvida com base no método de “seguir questões,” em vez de delimitar domínios distintos para a ciência e a política.

RESUMEN

Tanto la ciencia política como los Estudios Sociales de la Ciencia utilizan las palabras “política” y “ciencia,” pero cada una aplica significados tan diferentes que parecen casi inconmensurables. Aquí explico a los politólogos los distintos usos de estas palabras como son usadas por los profesionales de los Estudios Sociales de la Ciencia y cómo podrían relacionarse con los utilizados por la ciencia política. Muestro que sólo uno de estos significados de la palabra “ciencia” (de los cuatro posibles) la distingue radicalmente de “política” (que puede adoptar seis significados diferentes). Una vez definidos esos significados, propongo que podría desarrollarse una colaboración más fructífera entre los Estudios Científicos y la ciencia política basada en el método de “seguir las cuestiones,” en lugar de delimitar ámbitos distintos para la ciencia y la política.

KEYWORDS:

• Science Studies
• political science
• Bruno Latour
• science and politics

PALAVRAS-CHAVE:

• Estudos Sociais da Ciência
• ciência política
• Bruno Latour
• ciência e política

PALABRAS CLAVE:

• Estudios Sociales de la Ciencia
• ciencia política
• Bruno Latour
• ciencia y política

This article is related to:

Translating translation

1. The same question of political epistemology

Despite profound developments since the publication of Max Weber’s comparative study between German university professors versus state officials, Weber’s text continues to be a reference for summarizing the relationship between science and politics. Over the past century, however, numerous actors who increasingly blur the distinction between “scholars” and “politicians” have appeared. How should we, for example, classify researchers who disseminate their findings via press conferences? Environmental activists who are called as expert court witnesses against oil companies? Patent lawyers who establish camp within laboratories? A patient association’s board of directors which makes a choice between funding one or another line of research in molecular biology? A Ministry of Research that enforces the purchase of standardized protocol notebooks? A US governor who introduces into a state’s constitution the obligation to support stem cell research? A dissident physicist whistleblower acting against his own research institute? Members of Parliament who organize consensus conferences on controversial technology issues? IPCC [Intergovernmental Panel on Climate Change] climate experts voting to determine the likelihood of global warming causes – work that is then awarded a Nobel Peace Prize? Despite the proliferation of these new actors, a replacement for Weber’s classification – both slightly obsolete, yet at the same time apparently indispensable – is still lacking.

As with many other social science topics, one offered solution is to take this difference as an object of study instead of an indisputable resource. There is a history of successive distinctions between science and politics which can be documented so long as the distinction between the scientific and political spheres is suspended beforehand.

Although this program has a long history, its fruitfulness only became visible to political scientists after the publication of Steve Shapin and Simon Schaffer’s masterly volume on the relationship between Boyle and Hobbes (Shapin and Schaffer 1993). While previous scholarship had ignored Hobbes’ science and Boyle’s politics – or, at best, took one as being representative of science and the other of politics – their book seriously considered Hobbes’ role in science and Boyle’s in politics: or rather, it showed how the two traced the delicate and profoundly contested frontier between the cosmologies that would shape the experimental form of life that we associate with Boyle, and the new regime of authority authored by Hobbes.

Instead of rehashing the conventional history of the Scientific Revolution, the book opposes two cosmologies – two cosmograms (Tresch 2005). Boyle’s combines a particular literary style, a definition of God, a conception of the vacuum, experiments with an air pump, a definition of proof based on learned gentlemen’s opinions, and a particular conception of royal support. Hobbes’ includes a different definition of God, a different definition of the vacuum, a clashingly different definition of style and proof, a deep distrust of learned gentlemen assemblies, and as is well known, wholly different notions of authority, deduction, experiment, and censorship. In their accomplished work, the domains of science and politics are gradually dissolved and replaced by a recipe that defines increasingly irreconcilable cosmologies. These are the cosmograms that must be differentiated, even as they mix elements of the former “domains” of science and politics.

As the book shows, it was only much later, and by using nineteenth century definitions, that Boyle’s politics and Hobbes’ science could be forgotten and treated separately, making one the figurehead of experimental physics or chemistry and the other the symbol of political science itself – and before the serious question of “the relationships between the scholar and the politician” could be posed. Once the distinction had been made, however, few chemistry departments remembered Boyle’s political theology, and few political science departments remembered Hobbes’ wish to contribute to a finally settled science. By recapturing the entire episode of the Scientific Revolution, Shapin and Schaffer so-to-speak revascularized a thoroughly amputated question and proposed to use this distinction as an object of study.

Despite acknowledging the book’s importance, there doesn’t appear to be a truly productive debate between two related domains which otherwise use the same terminology: on the one hand, political science; and on the other, the study of the politics of science and technology known as the field of Science Studies.⁵ Both are nevertheless part of what could be termed political epistemology, understood as the analysis of the intricate associations between existing and potential conceptions of knowledge and of politics.

In the political epistemology debate that I aim to extend here, my first move will be to clarify an issue of terminology.

For starters, I am struck by how the combination of the terms “science” and “politics” evokes different reactions in each discipline. For political science, despite its many nuances, “science” appears to define a method, that is, a systematic, serious, rigorous, and empirical process to approach a topic: politics. While the boundaries can vary across the literature, defining them is possible, in principle (Grawitz and Leca 1985). The grouping of both words serves a similar function to that in other fields of knowledge such as “religion sciences,” “language sciences,” and “management sciences.” That is, parallel to other uses of the suffixes “-logy” and “-graphy,” such as in sociology or geography.

This is not the case in the more recent field of Science Studies. By translating the Greek expression “epistemology” into English, the sense of the two words “political sciences” and “politics” has been profoundly changed (Pestre 2007; Hackett, Lynch, and Wajcman 2007). While in “political science” the combination of the two words is not too problematic – it extends the social sciences, humanities, law, and administrative sciences to a new domain that is as suitable as the previous ones – this does not work for the other case: there, we face a powerful chemical reaction, if not an explosion.

Indeed, in the expression “the political history of sciences,” politics, as much as science, becomes a highly controversial topic within the same systematic line of research. That Boyle held ideas about kingship and Hobbes an opinion on mathematics should not surprise either historians of science or of political ideas. But Boyle’s conception of the vacuum profoundly affected what could be expected of physical matter and, consequently, of the social order. Hobbes’s conception of proof also modifies what could be expected of authority and, consequently, of the Sovereign. This undoes the possibility of aggregating scientific activities inside one sphere and political ones in another. It furthermore casts doubt on the ability of science to speak only about nature – because it is also about social order – and of politics to speak about sovereignty – because it is also about mathematical proof. We can’t even find comfort in saying that what may have been true in the seventeenth century is no longer true today. That a Nobel Prize has been awarded to the IPCC bluntly proves that all geopolitics is shaken by questions of mathematically modeling the Earth’s climate.

Political science does not need to question the legitimacy of its targeted domain – politics – despite the abundance of methodological and definitional conflicts (Favre 2005). The idea of a distinction between the natural sciences, the social science and politics (and more generally, “society”) is the specific object of study of Science Studies, but it turns out to also be its stumbling block (Latour 1999). Political science can turn to a grounded, often established, and stabilized version of the distinction between a method of inquiry (the science, or sciences) and a constituted object (the political), but it is precisely all the successive versions of this distinction which are the focus of Science Studies. Certainly, even if questions of political epistemology are common to both domains, political science’s relative calmness when uniting both terms is absent from the latter’s study of the politics of science.⁶

What I strive for, through this article, is to make a more equitable partition of both these concerns and this calmness. How can political science take as legitimate resources the ideas of “science” and “politics,” given these ideas are, at best, the provisional results of a history of science and politics that remain largely off-focus? Conversely, how does Science Studies intend to extensively “politicize science” without drawing from the grand well of political science for talking about politics? In principle, both fields should share a double concern: political science should be surprised that both its method and object can be questioned; Science Studies in turn should be embarrassed to call upon the word “politics” so often, without it ever having a clear meaning for it (De Vries 2007). On first approach, the intersections should be numerous enough for questions of political epistemology to finally form a fruitful locus of exchange.

In practice, the relations between the two domains are already plentiful, given that research topics commonly blend science, technique, and politics, and force one to simultaneously use concepts from both sides.⁷ Increasingly, the topics of doctoral students’ research that we co-supervise, the wording of conference invitations which we jointly respond to, and public concerns make our objects of study converge. The latter include ecological crises, urban problems, new information technologies, scientific controversies, instruments of globalization, the perception of risks, new military technology, etc. This means we often don’t know if we are moving within the domain of political science or the politics of science. And yet, it seems to me that despite these moves, our usages of “science” and “politics” remain incommensurable, or at least lack common effort.

Since I can’t speak in the name of political science, I would like to set the terrain of discussion and present to my colleagues how this minuscule problem of standardization could be presented from the point of view of my own discipline, Science Studies. This, however, can only be done by first specifying six different meanings regarding politics, and four regarding science – 10 in total, not counting possible combinations. I apologize in advance, as I would have liked to open the debate more succinctly, but the sole usefulness of this article comes from not artificially simplifying the matter at hand. The standardizer should not lay claim to cutting the Gordian Knot, but should know to be meticulous in the cut when it is needed. It is up to readers to decide if such “negotiations” should continue within such a tangled field … but, before deciding, it is worth reminding them that it is the contemporary world itself which is handing us these imbroglios so we can learn to untangle them.

2. Five (plus one) different meanings of the term “politics”

To establish the different meanings of the word politics as used in my discipline (and without much connection to its use in political science⁸), we need to dispel the idea that there could be a proper domain – politics – that is distinct or even separable, in principle, from all other domains. In a certain sense, we can accept the nevertheless-empty slogan that “everything is politics,” or at least that “politics is everywhere.” The reason behind this loose enunciation is that, most often, we study settings in which new beings, unknown up to that point, interrupt common life and thoroughly change the membership list of what I call, for this reason, a collective.

To use a famous example, when Einstein took pen and paper to address President Roosevelt to alert him of both the danger and advantages that would come from mastering nuclear reactions, was he doing politics or not? Certainly not in the same sense used by a specialist on post-New Deal US elections. Everyone would easily admit, however, he was doing politics in the most direct way possible, because from that point onward, atoms and their control would become part of military arsenals, would constitute one of the most salient geopolitical worries, would take up entire administrative divisions, and would worry new militant groups. The case is straightforward since Einstein was addressing a legitimately elected president through the most traditional lobbying channels. There is no great difficulty in extending the word politics to this case.

But, to understand what Science Studies really is about, the meaning must be extended ever so slightly, to the technical core of Einstein’s scientific work. Indeed, we no longer live in the same world, depending on whether we can, using the Lorenz transformations, make all reference frames (even accelerated ones) compatible; or whether in the absence of Special and then General Relativity, all reference frames are incommensurable, rendering motion impossible to calculate and thus making the universe incomprehensible. The issue in General Relativity is that the entire universe becomes knowable despite the dispersion of all points of view (Feuer 2005; Galison 2005a; Latour 1988). Is this, in the narrow sense, a political question? Clearly not, since it is one of the most esoteric aspects of physics. Is it then a strictly apolitical question? Clearly not, since first without, and then with relativity the number of accessible dimensions in the universe is altered. We literally can no longer move freely in the same world. Thanks to Marcel Mauss, we know that to interfere with frameworks of space and time is, undoubtedly, to touch the foundations of our common existence.

But if “everything is political,” how do we exit the confusion this empty slogan risks putting us in? We can no longer confine ourselves to a domain within which all “properly political” activities are gathered and clearly distinguishable from other spheres – economic, legal, scientific, etc. The solution I propose is to calibrate our somewhat contradictory definitions. This does not consist in defining a domain for politics beforehand, but to qualify, through consecutive terms, the successive stages through which controversial objects or issues – what the French call affaires⁹ – pass through. These multiple stages empirically define the contact points between questions that are common to our disciplines. In other words, I propose to start from cases and enumerate different ways to qualify their motions. I argue that, while clearly defining “politics” and “science” is impossible, it is absolutely possible to qualify the succession of stages through which controversial objects pass.

2.1. Politics₍₁₎ as new associations

Einstein’s example is not unique. The same is true of Pasteur’s microbes (Latour 2001a), the discovery of magnetism (Pestre and Néel 1990), the Généthon campaigns (Cardon et al. 1998; Callon and Rabeharisoa 1999), nuclear power in France (Hecht 2004), nanotechnology, financial equations (MacKenzie 2006), bird flu, etc. I consider the case to have been successfully made: the field of politics, in the usual sense, appears to be constantly overrun by the irruption of new beings that demand to be taken into consideration. For three centuries now, but ever more intensely nowadays, the immense majority of these new beings come from laboratories – in a broad sense that includes the social and cameral sciences.¹⁰

In another example, what would happen if by means of a new instrument we began discovering not a handful, but hundreds of planets outside our Solar System? For the scientists themselves, this is a “strictly scientific” and “clearly apolitical” activity. For Science Studies, quite the opposite, since a collective with or without exoplanets is not the same collective: it is fundamentally altered by reconfigured associations. One no longer lives in the same cosmos; in the same universe; or in the same one-planet world – our good old Earth – if there are an infinite number of planets, some of which can develop life. Moreover, if we hesitate to “politicize” such a technical question, we need only remember Giordano Bruno. This is still a burning question … and it will be the same case every time we modify the relations between members of the collective. Is the discovery of exoplanets political? Certainly not. Is it political? Yes, undoubtedly: a massive upheaval is introduced in the order of the world depending on whether we multiply or reduce their number.

From this, we recognize “political” within Science Studies in a first sense: that of a new association. It is probably much looser than what we would accept in political science proper, but it is nevertheless raw material for some technical and economic upheavals that public life will have to handle. But no, the sciences cannot calm, through their admirable rigor, the choppy waters of public life, with their involuntary (and at times voluntary) turbulence constantly disrupting the lists, ordinances, hierarchies, and the composition of beings assembled into a common world that is somewhat decent and livable.

This first meaning may seem paradoxical, since it appears strictly apolitical before the eyes of practitioners, the public, and some analysts, yet for us it is the most direct, universal, and violently political₍₁₎ meaning (I will add such clues to better distinguish each meaning, within this polysemy, throughout this article).

2.2. Politics₍₂₎ as raising the issue of the public

Here I introduce a second meaning of the word politics: do certain controversies force us to change the definition of the public? Clearly, the majority of new associations proposed throughout pages upon pages of patent databases, scholarly journals, and technical centers, disturb only a small fraction of people and will never become “political” in this second sense (although they are, as I have previously argued, political₍₁₎). For them to become political₍₂₎ it is necessary for the established forms through which collectives represent and administer themselves to become explicitly perturbed.

I’ll admit that the exoplanets issue has not become political₍₂₎ (at least not yet, unless the Inquisition Tribunal is still considering the matter …). No administrator, citizen group, or blog has yet posed the exoplanet issue as a subject of dispute that would challenge our institutional order, procedures, or usual methods of conflict resolution. In John Dewey’s terms, the public in this issue has not yet become, in the words of Walter Lippmann (2007), a problem.¹¹ Let us recall that for these two pragmatists, the public only emerges when the usual procedures that allow for the apprehension of unintended consequences of common action fail, one after the other (Dewey 2003). If the public appears, it is because the authorities no longer know what to do, and rules and procedures need to be re-invented (Zask 2000). At first “overshadowed” by the novelty of the questions posed, the community must then gradually recompose itself.

But for other topics, Dewey’s big question is clearly present, so that habitual forms of common life are deeply challenged. The case of Genetically Modified Organisms (GMOs) is perhaps the clearest: despite 15 years of procedural, legal, and regulatory innovations, the formation of an ad hoc public for this matter is still uncertain. More generally, daily news outlets multiply controversial topics, so that it becomes impossible to represent and gather the public under the same rubric for each single topic. The public – or rather, publics, in Dewey’s sense, and not that of the public space (Reynié 1998) – remains a problem, whether in the reintroduction of Slovenian bears into the Pyrenees, evolutionary models of climate change, or fishing quotas in the Iroise Sea (Marres 2005).

For those who want to continue separating science and politics into two separated spheres, an important issue could lead to a misunderstanding of my argument. A topic that becomes political₍₂₎ bears no resemblance to what scientists often call the undue, unjustified, and uncontrolled “politization” of a problem that “should have remained” “strictly technical or scientific.”¹² In my definition, there are no longer scientific and technical problems on the one hand, and political problems in the other, that at times get “unfortunately mixed up” and confuse good citizens. Further on, we will see how to index the three terms “politization,” “rationalists,” and “strictly scientific” in this habitual talk, which Science Studies has happily delivered us from. On the contrary, political₍₂₎ affairs are one of the possible responses to the innovations introduced by politics₍₁₎. As Dewey (after Lippman) brilliantly put it, the question of the public can only be grasped after impotence, destruction, and a breakdown of the usual forms of dealing with common life. It is as if certain scientific and technological innovations force us to compose the puzzle of public life with new pieces.¹³

2.3. Politics₍₃₎ as where the question of sovereignty is played out

There is a contrasting third meaning we can give to “politics” in Science Study cases. It is not about new beings who would break the habitual associations; nor about the device that could solve the question of a public shaken up by the irruption of new issues.¹⁴ Rather, it concerns subjects posed clearly before a collective, often by a State, about the very question of sovereignty and survival. Common sense will surely identify this: here, we have before us truly political affairs. Exoplanets, theories of relativity, GMOs: we don't really know what to do with them, but if you talk about war and peace, of friend and enemy, of domination and power, there we are in familiar territory.

In that case, it would be better to turn to Carl Schmitt than to John Dewey. Yet, these affairs being indeed political₍₃₎, they should not make us forget the other two meanings, applicable as they are to infinitely more cases and with trajectories that may never experience this third stage.¹⁵

What is special about the select cases that become politicized₍₃₎ in this fashion is that we find questions there that, while still scientific, involve issues of concern to all major political science questions. I have previously cited Einstein’s letter: it is clear the Manhattan Program would, in a matter of months and almost uninterruptedly, put the chain reaction problem – and with it, the enormous apparatus necessary for its industrialization – among the most burning issues of sovereignty, the darkest mysteries of esoteric power, and the most pressing questions of military artistry. For GMOs the situation will be in many ways the same, at least for Europeans and certain militant groups: it could clearly become one of those controversies where sovereignty is at play. Who owns rights to plant crops? Who owns living creatures? In a sense, it also becomes a question of life and death for peasants, for the planet, for farmers driven to suicide and imprisoned.

After the intensification of ecological crises, topics like climate, energy policy, urban planning, and others, have become outstanding issues of sovereignty, in which key decisions of being and not being are played out in front of a collective. How else can we understand that California defines itself, according to its governor, partly by what it can get out from stem cells and clean cars?¹⁶ Let’s not rule out the question of exoplanets ever becoming a political₍₃₎ question. If one day it became vital to leave a ravaged Earth, we would need to say that those who supervise the exodus are engaging, in the most radical way, in the great Schmittian question of decision-making. It might seem strange to associate questions of sovereignty with natural science but doing otherwise would limit the scope of political science far too much.¹⁷ “Friend” and “foe” have long ago ceased to refer only to humans, as plenty of entanglements require a fresh pair of eyes on the question of the Sovereign. The terms war and peace also apply to microbes … 

2.4. Politics₍₄₎ as challenges to democratic procedures

The contrast is even more salient considering a fourth meaning of politics, all the better attached to Jürgen Habermas rather than John Dewey. It concerns bringing matters enlarged by scientific and technical innovations into the regime of democratic discussion. As opposed to those that remain political₍₂₎ or political₍₃₎ – and a fortiori political₍₁₎ – it presupposes that such procedures already exist, that they are considered legitimate, that there is minimal agreement on how to deal with new cases, as well as a preconfigured acceptance by the minority for respecting the final decision. In short, everything that is lacking in the previous cases.

One may scoff at such an idealized conception of rational discussion since it is questionable whether profoundly technical issues such as GMOs or nanotechnology can be conveniently accommodated therein. A glance at the many existing examples of “public participation in scientific and technical decision-making” can justify a certain skepticism (Joly, Marris, and Hermitte 2003; Joly and Marris 2003; Fourniaux 2007). We might even predict that the participants in these exercises will tend to overflow the procedures, to reject selected representatives, and overall turn the issues into a “Dewey-like” politics₍₂₎ or “Schmitt-like” politics₍₃₎.

Yet, even if it goes unnoticed, every day multiple such cases are dealt with through such procedures, although they were formerly political in the previous senses. Entire administrative areas – those in charge of vaccines, fire prevention, mapping of risk-zones, pharmaceutical marketing, sale of pollution rights, and so on – lead to the application of legitimate procedures upon decisions that might, nevertheless, have controversial notes. Terms such as “politics of health” and “risk policy” designate such areas.

One might argue that there is little difference with the second and third senses of “politics,” since the same issue can drift across these four terms, but there is a distinct tone in the last one which distinguishes politics from what today is known as governance. Though they are all technical, they do not involve a questioning of procedures (the question of the public, in Dewey’s sense, does not arise or disappear). They do not lend themselves to the anxiety of the “state of exception,” and for this reason, they are classed as administrative issues. All these topics, coming as they most often do from laboratories, though highly technical (especially since law, management science, accounting, economics, and metrology usually add strands of their own to those of physics, chemistry, and biology), they nevertheless appear tractable to the interested parties. Succinctly, what is to be done in dealing with them is known; representatives are identifiable; the rules to be followed are known; validated information exists; procedures are known for where and how to speak. Only the final decision is uncertain. We can understand the attraction of politics in this sense and why we would want to extend it. While it is dangerous to reduce politics to governance, it’s not absurd to recognize that some affairs are well-politicized₍₄₎ in this sense.

If other meanings of politics have not always sustained productive debates in political science, the same cannot be said of the following question, common to both fields: does or doesn’t there exist a technical democracy (Jasanoff 2005)? The key question concerning expertise and that hybrid being, the expert, inhabits the meeting point between theories of democracy and the question of science. Can we extend the theories and habits of democracy to topics evermore loaded with expertise (Collins and Evans 2007)? In opposition to Weber’s famous book, this is not a question of confronting experts and politics, but of defining the type of politics₍₄₎ on which the controversial notion of the expert depends.

Oversimplifying, this overlapping domain is divided between those who think politics₍₄₎ can absorb technical issues without serious questioning, and those who, on the contrary, think it is science and technology’s political₍₁₎ and political₍₂₎ novelty that will always re-shape the collective, forcing new definitions of the public to arise – in other words, to turn it back to the second sense.¹⁸ The former think that politics₍₄₎ must be defined by procedure (in this case, of rational debate). The latter, that it can only be defined by the objects, the things, the “issues”¹⁹ that break it down every time (but, clarifying this point also requires to list the different meanings of the term “scientific”).

2.5. Politics₍₅₎ as governmentality

If we follow the future-facing paths of “issues,” we might conclude that many end up in what could again be called an apolitical state, were it not for the fact that this has been the focus of the finest analyses coming from political science, history, technology, and Science Studies. Indeed, the fifth sense will allow us to shed light on the previous two, by making use of Foucault to speed up the framing.

Using the rather burdensome term of “governmentality,” Foucault was able to detect a phenomenon that was equally weighty: the political₍₅₎ nature of what, up to that point, appeared to lie outside the political realm, such as prison layouts, sexual preferences, the enumeration of sins in religious orders (Quattrone 2004), law, glass ceilings in women’s careers, the definition of orgasm (Lloyd 2005) or masturbation (Laqueur 2004), and race-dog training (Haraway 2007a, 2007b). Speaking commonsensically, these issues do not appear to be part of the political arena, nor to be dealt with in the arena of democracy. Yet immense work by historians, feminists, and archeologists (in Foucault’s sense) has consistently revealed the forms of “power” that are exercised under the guise of professional, scholarly, and technical knowledges – the more effective precisely because they have become naturalized by blending into habits.

It is interesting for my calibration exercise to say of law or administrative habits that they are political₍₅₎ as saying exactly the opposite of what I started this list with. Far from breaking the associations of the collective, upsetting the hierarchies of beings that compose them, or violating the cosmos, like that which is political₍₁₎, governmentability exercises its violence through absolute silence, indisputable anchoring, and impressions of naturalness and obviousness successfully allocated to these institutions. Of course, this opposition is not so great in practice, since in either case it is the work of historians, feminist investigations, archeologists’ recovery of archives – and thus the work of other disciplines, laboratories, and researchers – that which enables us to see as politics₍₅₎ what was lost to habit and ignorance. It can be said that everything that is political₍₅₎ in a Foucauldian sense has also been previously political_(1),(2),(3), since most objects of governmentality have been topics of lively controversies when looked at through history of science, and have often violently involved questions of sovereignty. While today vaccinations are a silent part of administrative governance,²⁰ we can also look back at the raucous nineteenth century debates in Pasteur’s or Koch’s laboratories in collectively reorganizing themselves by making microbes their point of temporary survival (Evans 1987). It is surely around this tradition, roughly Foucauldian, that the collaboration between the two domains has proven the most fruitful (Meyet, Naves, and Ribémont 2004).

The cases we intend to follow are not always equally visible along their whole trajectories. While meanings _(2),(3) and ₍₄₎ allow for, if not easy, possible links to what our common sense recognizes as politics, this is not true for meanings ₍₁₎ and ₍₅₎. In both cases, one must make a great effort to see the eminently political character of what are, at first sight, “purely technical” questions. In the first case, it is extending the list of beings that upset the definition of the cosmos. In the fifth, it is a question of overcoming the overwhelming impression of obviousness through which the exercise of institutionalized power transits. These are two extreme forms of violence, and extremely effective forms of power/knowledge, which have long escaped the political arena before increasing their everyday roles in contemporary life. The question of how Science Studies and political science can work together to ensure the continued visibility of all these “issues” remains.

2.6. Several possibilities for the same cases

We can summarize the first results in Table 1. The multiple synonyms for the same word, politics, already evidence the extreme ambiguity – not to mention incoherence – of Science Studies. This is perhaps why political science has paid so little attention to such a confused field. Yet if we concentrate on the state of affaires and not on a priori definitions of what is “properly speaking” politics, the potential collaboration becomes obvious.

Table 1. List of different meanings of the word politics that can be used simultaneously by Science Studies.

Synonym	Definition	Benchmark
Politics(1)	New associations	Social history of science
Politics(2)	The problem of the public	Dewey
Politics(3)	Revisiting sovereignty	Schmitt
Politics(4)	The challenge of a democracy	Habermas
Politics(5)	Becoming an institution	Foucault
Politics(6)	That which is not scientific	Epistemology

The more so since one same affair, cause, or “issue,” can traverse all these different meanings. The motion will not necessarily be linear, but can jump stages, go up and down the table, or remain dormant in one category before moving frantically. If we take as a typical example of a political₍₅₎ subject the shape of the levees chosen by the Ponts et Chaussées engineers, it won’t be surprising to see that the choice of levee techniques to protect New Orleans suddenly becomes political₍₂₎ after the passage of hurricane Katrina; and even political₍₃₎ after President Bush’s catastrophic management: as he was unable to ensure the protection of his citizens, the obscure levee question became suddenly attached to the grand issue of sovereignty. A spatial translation will have the same effect as a temporal one: we will find across the world, in all political states, apparently similar “causes” for local views on water supply, public transport, sewage, the protection of species, renewable energy, and humanity – displacement transforms the same person from a “citizen,” for an institution South of the Rio Bravo, into an “illegal alien,” for one North of the Rio Grande.²¹ I believe that, to deal with these issues collectively, all we need is a bit of coordination. To wit, we also need to notice that none of the meanings that I have put forth entails, so far, a distinction between other domains and that of “the sciences.” All of them refer to the same common assemblage and only qualify stages, so to speak, in the cooking of the ingredients that make up this melting pot.

This will not be the case for the sixth sense of the word politics. To define it, we will have to extend our clarification efforts and now deal with the word “science.”²²

3. Four different meanings of the word “scientific”

It may appear foolish to try and pin down the use of the polysemic word “science” in just a few pages. Nevertheless, it would be impossible to calibrate the collaboration between Science Studies and political science without trying to capture the different meanings of the term, since it is precisely the point of our field to make its use as problematic as possible.

However, if there is one term that the social sciences should not use without care, because Science Studies has already extensively scrutinized it, it is the word “science” (although I will ignore the word “social” here, which if anything is more problematic, as it involves an altogether different discussion … see Latour 2006). The important point is to somewhat polemically recognize in this opening move that “science” is not that which clarifies, but what needs to be clarified. For our discussion, it must shift from being an indisputable resource, to becoming a topic of inquiry. In other words, for the discussion to take place, we must first accept that we have no idea what the term “scientific” means – which, of course, you will realize is the counterpart of what I initially asked for the word “political.” There is no more a predefined domain of politics than there is a method, domain, discipline, or sphere that is “scientific.”

3.1. Scientific₍₁₎ as rational and objective

I will quickly go over a first sense of the word scientific₍₁₎ (using again indices), which is the one taken as a rather vague synonym of “reasonable,” “rigorous,” “honest,” “serious” – a rather ill-defined mixture of good moral qualities and “gentlemanly” habits. This is not to say the term is not important: quite the opposite. Increasingly, quasi-ethnographic and even ethological studies of “scientific” behavior, of “scientific” bodies themselves – even regarding scientific feeding habits! – show to what degree this package of qualities falls under the “scientific” rubric, without it ever being made completely evident. It is a way of being, of dressing, of behaving, of taking distance, of being attentive, and if we could summarize it in one expression, of manifesting some variant of the pathos of objectivity (Daston 1992; Shapin 1994; Shapin and Lawrence 1998). A pathos which is likelier to be found under the three-piece suits of state advisors than under laboratory white coats (Latour 2002). But this very special meaning, even if it is of great interest to anthropology, has little to do with political science’s interests and is more a matter for a history of characters. Let us only recall that, in this trivial sense, to say of a statement that it is “scientific” means nothing more than it is “trustworthy” without explicating the chain of reasonable people who can vouch for it. Just as “objective” sometimes designates qualities of subjects and sometimes qualities of “objects,” so the adjective “scientific” changes its meaning completely, depending on whether it is used to designate a frame of mind or a relationship with a particular type of object.

3.2. Scientific₍₂₎ as that which is different from politics

The second meaning brings something to the foreground which is of definite interest for us, since it deals with the very invention of a distinction between the domains of politics and science. Science Studies can adequately be defined as the field which, instead of taking the existence of this distinction for granted, has made efforts in the opposite direction: to describe those historical episodes when it is invented. Rather uncomfortably, instead of resting on either side of the wall that separates science and politics, Science Studies rests its tools upon the wall itself, even while it is being built up.

As I described in the opening sections, the most striking case, even in France itself (and even for political science), is that of the relationship between Boyle and Hobbes as described in Steven Shapin and Simon Schaffer’s masterpiece. We can, nevertheless, go back further and look not only at the “science” and “politics” debate – a three century-long history which is now becoming clearer – but up to the very origin of the crucial difference between conviction and persuasion. Another recent masterpiece, by Reviel Netz (2003), allows us to again historicize this older differentiation, which from the Scientific Revolution then inherits.

Where does the astonishing idea that still excites readers of Plato’s Gorgias – that in addition to rhetoric, there is another path, that of indisputable proof – come from? Using terminology borrowed from Barbara Cassin (1995), from where does the decisive difference between terms at first synonymous – that of apodeixis, which will result in apodictic reasoning, and that of epideixis which will give eloquence and deceptive appearances – come from? In backtracking to the source, our endpoint is what will later turn into the separation between the domains of science and politics. Which sensible reader might remain unconvinced of the essential, indisputable, indispensable difference – both intellectual and moral – between a rational and scientific discussion on the one hand, and on the other, an open-ended debate which must be closed through the resources of rhetoric?

It is precisely the great accomplishment of Netz’s book that it brings us, through work that is as rigorous as that of Shapin and Schaffer, closer to the “laboratory” from where this distinction will emerge. Succinctly, apodicticity is the transposition into philosophical language of Greek geometers’ practice – who only succeeded in transferring “necessity” from one end of a proof to the other because they stuck to the narrow framework of formal geometrical diagrams with obsessive and jealous watchfulness. The idea of a radical difference between the apo- and epideictic comes from the Platonists’ hijacking the use of esoteric practices they never succeeded in imitating (geometry “worked” because the diagrams had no content). Indeed, Platonists only managed to relocate the stunning rhetorical effect: the possibility to transfer, step by step, indisputable certainty. Without going into the details, this puts us at the very heart of (political) epistemology (Latour 2008). Platonists reinvented the life of the City by importing a radical novelty, which was empty except for the endless reiteration that there is a radical difference between philosophy and sophistry, between apo- and epideictic … this is how reason would, apparently, triumph over sophistry.

If we say that a statement, a fact, a discipline, or a person is scientific in this sense, we never define a particular content. We simply draw the difference between the inside and the outside. We trace out a territory. We reject something that, later, will become the proper domain of politics defined precisely as what is not scientific. This is where political epistemology becomes epistemology in full.²³

It is this operation of sidelining that will force us to define a sixth meaning of the word politics₍₆₎, referring exclusively to that which lacks the rational qualities we imagine the sciences to have. Henceforward, we will act as if any idea of politics depended on this definition of science, and vice versa. Henceforward, the two definitions form a damned couple that can neither live together in agreement nor in divorce, but whose identity is based on the fact of being radically separate (inseparable) from the other … “Ah, it’s scientific₍₂₎? So, it’s not political₍₆₎!” It is a staggering term with the sole objective of excluding one domain from the other, with no other content than this very exclusion! But it is a term of staggering effectiveness since, without exaggeration, one can draw a continuous line from Gorgias to our present day: the weapon of epistemology is hardly blunted, but rather more as if twenty-five centuries had continuously sharpened it!

These two usages of scientific₍₂₎ and political₍₆₎ were obviously used in the past, for example, in Louis Althusser’s distinction between “science” and “ideology.” More surprisingly, it is still these terms that we use to police the border between “purely technical questions” and the tendency to “unduly politicize₍₆₎ what should never be politicized,” even though today’s affairs are totally outside both this idea of science₍₂₎ and this notion of politics₍₆₎. No wonder the difficulties we have in coping with the situation!

To put it more bluntly, common sense generally recognizes in politics₍₆₎ only what is excised by the epistemological scalpel; only the negative, of which the scientific₍₂₎ is the positive. If, by misfortune, political science was to take up this sense of science in its name, it would clearly have some difficulty digesting it. This would imply in its usage, and in the definition of its method of investigation, a polemical term par excellence that would make its object of study – politics₍₆₎ – a mere residue!

But the most fundamental point I have already alluded to and to which I will return in the conclusion is that, of all the meanings of science that I put forth, it is this second one – and only this one – that is exclusively devoted to distinguishing itself from politics. Of the five synonyms I have listed in the previous section, not one of them seeks to create a domain of its own, that of politics. All these terms refer to different stages in the trajectory of the same affairs, much as astronomers have taken to using different names (red dwarf, supergiant, super supergiant, supernova, pulsar, black hole, etc.) for various states of the same stars. The purpose of all these terms is not to draw a line between that which belongs to scientific objectivity and that which depends, unfortunately, on political carping, but to define how the collective manages to relatively disentangle itself from these matters.

3.3. Scientific₍₃₎ as tested in front of spokespersons

It is exactly onward from this limit that the third meaning of the word “scientific” arises, when the term becomes the object of a meticulous investigation instead of the In hoc signo vinces²⁴ on the eve of battle. Arguably, Science Studies became constituted when the proposal arose to define scientific₍₃₎ activity precisely without invoking the epistemological question par excellence, i.e. without accepting that its research be mobilized in the great war of science₍₂₎ against politics₍₆₎ (Pestre 2007). The misunderstandings that this field initially gave rise to stem from this: some researchers thought that we were taking the initiative to politicize₍₆₎ the natural and the social sciences, when in reality it was a matter of studying them positively by preemptively ignoring the question of demarcation. Hence the battles between “realism” and “relativism,” which made sense to those who held the labarum at arm’s length but were about as senseless to us as those between Big Endians and Little Endians.²⁵ It was thought that we were going to war against Reason, whereas we were, on the contrary, deserting its fight and melting our swords to forge ploughs!

Freed from the second meaning (and therefore from its cumbersome political₍₆₎ counterpart), the adjective scientific₍₃₎ does not refer to the same class of phenomena at all. First, it directs us towards new places such as the laboratory, which it perceives as a type of practice, its anthropological strangeness then appearing in full light (Houdart 2008; Knorr-Cetina 1999). Surprisingly, this practice had rarely been described before, and the objects that are revealed differ completely from the idea that we usually hold of science. It is not the purpose of this article to summarize its content, but I would like to point out one of the ways it could act as a hook for the concerns of political science.

In this sense, science₍₃₎ is any device that makes it possible for phenomena to speak, and for which the researcher becomes, in a way, the spokesperson. With the immense literature and the extreme difficulties in defining, within political science, notions such as representatives, intermediaries, brokers, translators, and interpreters, we can envisage potentially ample collaboration with the increasingly rich literature describing how researchers represent their objects (Lynch 1985; Rheinberger 1997; Atten and Pestre 2002). The fact that one of the meanings of “representation” applies to humans in a traditional political sense (the Hobbesian side), while the other meaning of the word “representation” applies to things in a traditionally scientific sense (the Boyleian side), allows that any controversy, in practice, has already mixed up the two forms (Stengers 1993).

“In whose name are you speaking?” “Are you a reliable witness?” “Is it you speaking, or those in whose name you speak?,” etc. Clearly these are common questions, whether in global warming, the calculation of unemployment statistics, or the representativeness of trade unions. To avoid seeing what is common to both science₍₃₎ and politics, it is sometimes claimed that, in the former case, it is “the facts that speak for themselves”; whereas in the latter, at the end of the day it is only humans who speak “in the place” of other humans. But, first of all, it is no easier to make humans speak than objects and, second, facts never speak for themselves without an infinitely complex device to make them speak (Daston and Galison 2007). The common question is rather to know which trials²⁶ are put forward and how the stubbornness of the objects we talk about is ensured, depending on the case. Whether one is doing demonstrations with rats or street “demonstrations,” one must always demonstrate. And if you demonstrate, you will always be contradicted. You will always have to find the appropriate instrument, delimit the right arena for the dispute at hand, summon a group of witnesses capable of making judgments, and specify the trial that will make the difference, then imagining the procedures by which this can be closed.

Moreover, there is some irony in writing an article to convince academic colleagues to calibrate our definitions, while on every side, devices are multiplying that are thirty years ahead of our vocabulary and which have already, in a thousand ways, settled in practice the question that still scandalizes us! When the IPCC met in Paris in February 2007 to serve as an ad hoc assembly on the issue of global warming and to decide on its precise causality, what use could we have in polarizing the resources of political science and Science Studies? It is clear that efforts to fit such an assembly into the Procrustean bed of opposition between true science₍₂₎ and false politics₍₆₎ – even though they have not been lacking – are no longer meaningful. What we are already heading towards is a completely different issue: which epideixis can lend support to apodeixis?

Let us note that no feature of this anthropology of science differs radically from the first five meanings of the word politics that I have identified. Of course, there are nevertheless differences: one does not conduct a laboratory experiment in the same way one conducts a strike or leads a board of directors or a committee of experts. But there is certainly not the difference imposed by the previous meaning. The skills, the careers, the issues, the passions are all different: the cauldron is common, as are the affairs, and the “issues” that simmer within. This is the common object, the res publica (Latour and Weibel 2005). Whereas with the previous meanings, it was impossible to conduct an investigation by following the definitions’ leading issues because of the weight of the definition of what is scientific₍₂₎ (and its hindrance by politics₍₆₎), there is nothing to prevent us from now hopping from the most esoteric laboratory to the noisiest of assemblies. It is not the old frontier between science and politics that must be scrutinized, but successive stages of each affair.

3.4. Scientific₍₄₎ as data logistics

If we hesitate to see the extent to which scientific₍₃₎ activity need not be distinguished from political situations in order to be understood, it is perhaps because it is also confused with a fourth meaning – as different from the previous (third) as that one is from the second. If you confidently assert that a statement is scientific₍₄₎, it is often as the iceberg’s tip of an underlying, sizeable collection of data. In opposition to the first meaning, what reassures you is not that the person making the statement is a scientist₍₁₎, or that he or she appears “serious,” “reasonable,” “distant,” or “dry.” It is not their subjective qualities or their behavior that you are referring to. Indeed, what supports your following them is that you have been able to gauge, often indirectly, that there is a continuous flow of calibrated and archived forms – data, or “givens” – to support their claims. In fact, I propose to call these objects, given their costs and the difficulty of obtaining them, “crafteds” (sublata).²⁷ Scientific₍₄₎ then means that one is supported by evidence which one could go back to, if challenged, by tracing a seamless path.

The Internet has made access to databases so widespread that anyone can now check for themselves what it means to make a statement based “on data.” But we should not forget that this is a phenomenon that goes back several centuries. The digitization of databases partly facilitates fluidity and access (despite the opposite trend of secrecy²⁸), but it merely extends what was already being done through collections, archives, prospecting campaigns, and statistical surveys. Every scholarly discipline, be it geology, obstetrics, ethnography, pedagogy, management, tribology, political science, etc. will be defined by the size, cost, quality, and maintenance of its network of databases. It will be more or less scientific₍₄₎ depending on the size of this network.

This, nevertheless, does not mean that it will be scientific₍₃₎! No one would judge a pension fund by the size of its capital: it is judged first and foremost by the return it has been able to give back. Similarly, it would make no sense to judge a discipline as scientific₍₃₎ on the pretext that “it has a lot of data.” It must first show us what it can do with its capital, and how it has managed to make it bear fruit by producing recalcitrant objects. Here again, we must not confuse the entirely different meanings of the dangerous adjective “scientific.” There are multitudes of scientific₍₄₎ fields with mind-boggling amounts of data that have not produced a single scientific₍₃₎ result that can be trusted, even though they have managed to inflate the importance of those who borrow from science₍₁₎ all the pathos of objectivity. If there is one thing that Science Studies have liberated us from is the belief that it is easy to imitate the scientific style and that it is enough to be boring to be saying scientifically accurate things … one thing is certain: you can’t use the word “scientific” without further specifying the relevant clues of usage.

My hope is that adding some clues to untangle the synonyms of the word “science” will help the social sciences to refrain from being intimidated: they will look very differently as either social sciences₍₃₎, or as social sciences₍₂₎. Above all, their political impact will be incomparable. Émile Benveniste was surprised that the genius of the French language had not derived from the Latin word scientia, terms like “sciental” or “scientaire,” and instead just stuck to the single word “scientifique.” Such a situation would have allowed us to keep terms separate whose distinction seems essential to our common life.

4. Conclusion: cosmopolitics

By summarizing in Table 2 the different meanings identified in this article, we can see even more clearly that just one of these meanings (in bold) is defined in contrast with that in the other column. This is the only meaning that can be said to really trace incommensurable spheres. All the others designate, more or less precisely, different (and sometimes successive) states in which the matters dealt with collectively can be found.

Table 2. Summary of the different meanings of the words “political” and “scientific.”

Synonyms of	Political	Synonyms of	Scientific
Political(1)	New association
Political(2)	The public and their issues
Political(3)	Sovereignty
Political(4)	Rational discussion
Political(5)	The invisible institution	Scientific(1)	Reasonable
Political(6)	Non-scientific	Scientific(2)	Non-political
Political(1)		Scientific(3)	Product of trustable witnesses
		Scientific(4)	Mobilization of data

This is a very rough sketch of the delicate question of the relationship between “science” and “politics” from the perspective of Science Studies (at least as I interpret it). The nuances I have introduced may appear, to the eyes of political science, both too nitpicky and too rudimentary. It was important for me to distinguish between definitions that make collaboration possible and those that negate it from the outset, and hence the importance of proceeding slowly.

When we use both terms too quickly, we borrow without thinking the only definition of the word scientific₍₂₎ that has no content other than to define politics₍₆₎ by default. All other meanings are not intended to define distinct domains or autonomous spheres of activity and, all at once, the collaboration with disciplines such as political science and science policy either becomes completely impossible or consists of taking as a starting point the two caricatures. This must then be patched up: there are “also” political “aspects” or “dimensions” in the sciences. Despite the extent of the commentary on Weber’s book, we must insist that, on the contrary, there is never any connection to be made “between the scholar and the political,” or science and politics: for the simple reason that these fields have never existed separately.

In this article, I have tried to propose a calibration of terms to define a phenomenon as we observe it from my chapel, that of Science Studies. I am convinced that there is common ground with political science, even if our different uses of the terms “science” and “politics” don't always make the superposition of cases easier. In other words, it seems to me that the real phenomenon that we are trying to trace is not found in either of our two disciplines, but rather in the more or less rapid evolution of cases, the successive stages of which must be very precisely qualified. “Science” and “politics” do not denote domains of activity or competence, but rather states – like liquid or gas states – in which the controversial objects that make up the common world are to be found. I believe it is the precise definition of these states that the present work can add to, in what could be called the politics of science and the various sciences of politics.

Science Studies would surely benefit from such a collaboration, as our use of the word “politics” – let alone “social” – is so confusing. Perhaps political science could also benefit from a different way of talking about the sciences₍₃₎ in order to rediscover the links with the world and with the cosmos, since this is what it is all about. Hence the general term cosmopolitics,²⁹ which should be chosen to cover all the terms in Table 2 – all but one. The scale of the current ecological crises makes it increasingly easy to understand that all politics has always been a cosmopolitics, i.e. a politics of the cosmos and that, in this sense, we have always been dealing with what I have called an object-oriented politics (Latour 2005). This is a considerable task, I readily admit, but it is no longer unfeasible.

While, for so many centuries, we hoped to rescue unfortunate politics by the apodictic reasoning offered by one or other science₍₃₎ (theology, physics, biology, economics, genetics, systems theory – there has been no shortage of candidates) we see today that, on the contrary, it is politics, but a very different politics from the one that had been despised for so long, that comes to the rescue: to serve as an asylum and shelter for the positive sciences₍₃₎. Basically, this is nothing less than to uninvent the Platonic solution which, in trying to set up politics in reason, ended up losing both science and politics. What a reversal! Is it so absurd to think that political sciences and Science Studies could contribute, by joining forces, to the elaboration of an alternative?³⁰

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Bruno Latour

Bruno Latour (22 June 1947–9 October 2022), political/social scientist and philosopher, was one of the most influential scholars in Science and Technology Studies (STS). From groundbreaking early works, such as Laboratory Life, to conceptual/methodological frameworks such as Actor-Network Theory (ANT), and finally his focus on political ecology in his latter years, Latour's work spread far beyond STS's borders – cultural, linguistic, academic and geographical – to become familiar across a diverse spectrum of fields in the social sciences and humanities.

Notes

5 Translator's note: “Science Studies” in the original French manuscript, and in subsequent appearances.

6 See thematic clusters put together in two leading journals in the field: Social Studies of Science, particularly Volume 37, October 2007, and Science, Technology and Human Values.

7 To provide a few examples as testament, and sticking to France, see among others, Lolive (1999), Torny (1999), Callon, Lascoumes, and Barthe (2001), and Gilbert (2003).

8 For an example of such incommensurability, see Donegani and Sadoun (2007), where no definition overlaps with the ones proposed here.

9 Closely following the sense laid out in Boltanski et al. (2007) and more generally, in the works of Luc Boltanski and the Groupe de Sociologie Politique et Morale (GSPM).

10 Translator’s note: see Desrosières (1993).

11 See furthermore Marres (2007).

12 The same point is made in Lagroye (2003).

13 On the notion of “progressive construction of a common world,” see Latour (1999).

14 Translator’s note: English “issues” also in the original.

15 This is the focus of the case by G. de Vries, which shows how an issue can, regardless of efforts deployed, never pass that state.

16 “Sect. 4. Article XXXV is added to the California Constitution, to read: There is hereby established a right to conduct stem cell research which includes research involving adult stem cells, cord blood stem cells, pluripotent stem cells, and/or progenitor cells. Pluripotent stem cells are cells that are capable of self-renewal, and have broad potential to differentiate into multiple adult cell types. Pluripotent stem cells may be derived from somatic cell nuclear transfer or from surplus products of in vitro fertilization treatments when such products are donated under appropriate informed consent procedures.”

17 See the amazing imbroglio drawn up in Barthe (2006), where the issue of nuclear waste changes even the notion of decision as a topos of sovereignty theory.

18 This is the distinction introduced by M. Callon, Lascoumes, and Barthe (2001) between “delegative democracy” and “dialogical democracy,” the latter becoming necessary because of the limits of the former in all controversial issues.

19 Translator's note: English “issues” in the original.

20 Translator’s note: this, of course, has changed remarkably in the post-COVID-19 world, lending even more strength to Latour’s idea that politics/science discussions can deeply change their interpretative loci up and down the political classification proposed here in times of crisis (see below); or even that these changes in and of themselves define times of crisis.

21 Translator’s note: in the original, Latour drives the coda of this sentence with an impossibly difficult metaphor referring to sovereignty across and under the Alps. It seemed to me that, with its publication in Tapuya, the definition of humanity across migratory borders would be much easier to grasp.

22 This is in contrast to Favre (2005), which takes as its starting point that in order to discover the epistemological “foundation” of political science, it is necessary to first define a “scientific worldview.”

23 Care must always be taken to distinguish “political epistemology,” which encompasses studies of the origin and management of differences between science and politics, from “(political) epistemology,” which encompasses studies of the origin and management of differences between science and politics. Under the pretext of defining science, “(political) epistemology” puts in brackets the fact that it is all about politics and nothing else. See Latour (2001b).

24 Translator’s note: “In this sign thou shalt conquer.”

25 Translator’s note: though Latour here probably referred to the two Lilliputian tribes in Swift’s Gulliver’s Travels which argue over the proper way to peel a hard-boiled egg (whether by starting at the big or smaller end), the terms have also been adopted by computer scientists in similarly obtuse debates regarding the “best” way in which information can be organized for storage. See Cohen (1981).

26 Translator’s note: There is a pragmatist connotation here in the original French épreuves, which refers to experimentation oriented to settle controversies and which allows to stabilize what is conceived as “reality” until the next time it is put to the test.

27 Translator’s note: introduced as English/Latin data in the original, paired to the French données, which can be literally translated as “givens.” Latour drives hard the point that data/données are never passively “given” but are instead appropriated after a process of manufacture. Latour therefore proposes the French word obtenues (“obtaineds,” “takens”) as more appropriate terminology, in opposition to donnés/data. The sublata paired to obtenues is the vocative case of the Latin sublatus, which refers to having been (1) “raised, lifted or elevated,” (2) “removed, or taken away,” and (3) “destroyed or abolished.” I have instead chosen to translate “obtenues” as “crafteds” as a reflection on Latour's point that “data” (givens) require enormous amounts of skilled labor to be created. For the idea of data-production and its relation to data analysis as two forms of craftwork, particularly in data intensive sciences, see Bartlett et al. (2018).

28 See Galison (2005b).

29 In the sense of I. Stengers (1996a, 1996b), but also in the sense of Beck (2006).

30 I am very grateful to Dominique Linhardt and Dominique Pestre for their comments on this paper.