Archeology,genealogy,and geography of experimental philosophy

The air-pump as an archeological site

In the internalist historiography, the air-pump produced an immutable fact – the existence of vacuum in nature. In presenting Boyle’s experiment as a canonical case history in the development of modern science, James Bryant Conant (1970) defined science ‘as a series of concepts or conceptual schemes arising out of experiment and observation [that constitute scientific facts] and leading to new experiments and new observations’ (p. 4). According to this long-naturalized assumption of universal and timeless nature that allows for transparent experiments, immutable facts, truthful representation, and linear scientific progress, the existence of vacuum should have been an automatic conclusion from the proliferating, ingenious, and presumably transparent experiments especially by Blaise Pascal. Not so, according to Boyle, who understood why the Plenists and the Vacuists could only talk past each other, given their irreconcilable definitions and limited instrumental control. He refused to engage in the ‘metaphysical’ question of whether a ‘truly empty’ space exists and redefined vacuum as a ‘physiological’ space produced by the air-pump. This instrumental concept (see Ihde, 1991; Kim, 2008) of vacuum demarcated experimental philosophy based on the control of nature from the traditional natural philosophy that pursued speculative truth about nature (McCormmach, 2004). What counted as philosophy had to change fundamentally to establish the cognitive/institutional space of experimental or ‘new’ philosophy.

In the authors’ informal use of Wittgenstein (Bloor, 1983), doing experimental philosophy required a new ‘form of life’ or ‘language-game’ (p. 15). In order to place the air-pump in its proper context, we must ask: What was natural philosophy? Who were natural philosophers? What did they seek to accomplish through natural philosophy? Shapin and Schaffer’s strategy was not to explain the natural by the social, as internalists have charged, but to configure the coproduction of their ideal forms under specific conditions. They forced us to rediscover the basis on which particular forms of knowledge became possible in Restoration England. Leviathan and the Air-Pump can be seen as a project of Foucauldian archeology which scrambled our neatly drawn disciplinary boundaries and linear (whiggish) histories to configure an ‘epistemological space specific to a particular period’ (Foucault, 1973b: xi) and, as I point out below, a particular locale. By directing our attention to the pre-disciplinary configuration of knowledge-producing activities, Foucault wished to discern a ‘well-defined regularity’ or the ‘rules of formation’ that defined the objects and shaped the concepts and theories of empirical knowledge (Foucault, 1973b: xv–xxiv, 1972: 178–195). His archeology fixed our attention on the level of what made knowledge possible (or the rules of structuring), rather than what it became (or the structure).

By focusing on the controversial process of stabilizing this interdependency between the air-pump and the concept of vacuum, Shapin and Schaffer treated the air-pump as an open archeological site to unearth the conditions under which a machine could be trusted to reify philosophical concepts and practice. In other words, they questioned its self-evident importance in the conceptual history of modern physics and made intelligible to us why it was so difficult to settle the notion of vacuum in Restoration England. In order to write a history of madness, Foucault (1988) had to ‘renounce the convenience of terminal truths’ or avoid guidance from the modern characterization of this ‘tragic category’ in the language of psychiatry, a discipline that has produced ‘a monologue of reason about madness’. His ‘archeology of silence’ was meant to recover the ‘broken dialogue … all those stammered, imperfect words without syntax in which the exchange between madness and reason’ had been made (pp. ix–xii). Only in light of Hobbes’ alternative discourse, Shapin and Schaffer similarly argued, could we appreciate what was at stake – the very goal of natural philosophy and its function in forging legitimate polity – in using a fragile machine to produce an item of philosophical knowledge. By recovering the content of his natural philosophy that had been relegated to oblivion by the success of modern physics, they developed a historical memory of the conflicts and struggles that shaped Restoration science.

Shapin and Schaffer limited their analysis to the particular locale of Restoration England, which did not follow Foucault’s pan-European scheme of the classical episteme for the 17th and 18th centuries. Even so, their emphasis on the distance between Hobbes’ natural philosophy and Boyle’s experimental philosophy may be considered an exercise in Foucauldian archeology or ‘general history’ (in contrast to ‘total history’) designed to take discontinuities seriously to eschew teleologies and totalizations (Foucault, 1972: 3–17). They shared his critique of the internalist history of mathematical and physical sciences as ‘the almost uninterrupted emergence of truth and pure reason’ (Foucault, 1973b: ix) and his insight into the rupture and reconfiguration of the episteme, no doubt reinforced by Thomas S. Kuhn’s (1962) work. ³ In delineating the transformation of the classical episteme that exposed its limits of representation, Foucault emphasized the ‘total’ difference (Foucault, 1973a: x) between natural history and biology. ‘Biology did not exist’ until the 19th century, he put bluntly, not just because the disciplinary term did not exist (J.B.L. Lamarck coined ‘biology’ in 1802), but because ‘life itself did not exist. All that existed was living beings, which were viewed through a grid of knowledge constituted by natural history’ (Foucault, 1973b: 128). Similarly, experimental philosophy had to invent a set of practices that made ‘something that does not exist able to become something’ in order to escape the interpretive framework of natural philosophy intent on fixing the (speculative) truth about nature. For the vacuum to exist, it required a new ‘regime of truth’ (Foucault, 2008: 18–19). By stabilizing an epistemic/social/political space in which an artifact could be seen as a fact of nature, the Royal Society constituted a new space of representation that legitimated their right to determine the truth of nature. In the process, the fragile air-pump became a powerful emblem for a parliamentary order. By reading Hobbes’ Leviathan as natural philosophy and Boyle’s experimental philosophy as a political project, Shapin and Schaffer not only unearthed competing forms of Restoration knowledge, but also raised the possibility that all legitimate knowledge has to interweave moral, political, and philosophical concerns. Philosophers have responded to this challenge with their own musings on social epistemology, historical epistemology, and historical ontology (Fuller, 1988; Hacking, 2002; Rouse, 1987; Tiles and Tiles, 1993).

A genealogy of facts

Despite externalists’ enthusiastic endorsements, the divergent knowledge products Hobbes and Boyle developed under a shared context did not stipulate a causal relationship between science and culture. If they shared sociopolitical context, matters of concern and even the mechanistic ontology of nature, what set them apart? What were the rules of their philosophical games? In order to explain their differences, Shapin and Schaffer focused on how they sought to manage dissension through specific methods of knowledge production. In doing so, the authors appropriated for the history of science Foucault’s genealogical method that seeks to identify stabilized techniques of power–knowledge regimes. They displayed scientific method ‘as real practical activity’ or ‘as crystalizing forms of social organization and as a means of regulating social interaction within the scientific community’ (p. 14). By prescribing how to do philosophy via the certain method of geometry, which promised incontestable knowledge by starting from agreed-upon definitions (rather than opinions) and proceeding with reason (instead of private belief), Hobbes sought to avoid another civil war that could ensue from religious factions and scholastic debates. Dissent was never safe. In contrast, the Royal Society deployed an ensemble of material, social, and literary technologies to institute experimental philosophy as a parliamentary production of consensual knowledge. A social body was ‘the effect … of the materiality of power’, as Foucault (1980) observed, rather than that of a consensus deriving from ‘the universality of wills’ which he regarded as a ‘great fantasy’ (p. 55). The book thus tackled, as Richard Westfall (1987) recognized, ‘a major historical question: what was the connection between modern science and the new social order in the West that came into being at much the same time?’ His protest – ‘Who will accept the proposition that Boyle fathered experimental science as part of a dispute with Hobbes about the right ordering of society?’ – testifies to the complexity of the historical project in which Shapin and Schaffer invited us to participate. A sudden epistemic change in Foucauldian genealogy takes place not necessarily because of individual actors’ intentions, interests, and dispositions, but through a displacement of the object on which the technologies of control operate to engender an entirely new configuration of power–truth–knowledge complex.

If Foucault’s (1979) premise on ‘power–knowledge relations’ –

that power produces knowledge; that power and knowledge directly imply one another; that there is no power relation without the correlative constitution of a field of knowledge, nor any knowledge that does not presuppose and constitute at the same time power relations

– was not new, his approach to exposing their intimate, embodied relations that structure our lives and thoughts was. Instead of analyzing these relations through historical actors and their particular interests and dispositions, Foucault proposed an analysis of the microphysics of power or an engagement with the ‘body politic’, defined as ‘a set of material elements and techniques that serve as weapons, relays, communication routes and supports for the power and knowledge relations that invest human bodies and subjugates them by turning them into objects of knowledge’ (pp. 27–28). Power and knowledge are coproduced and maintained by the techniques of disciplining the body. Just as Foucault (1979) focused on the prison, its material organization and its machinery of disciplining the body to undermine its ‘self-evident character’ (p. 232) for an embodied ‘history of the modern soul’ (p. 23), the authors questioned the self-evident status of the laboratory in modern science to expose its constitutive and operational requirements and its legitimation of machine-based experiment ‘as a privileged means of arriving at consensually agreed knowledge of nature’ (p. 3) for a technological history of modern science. The prison and the laboratory can be seen as power–truth complexes that forged modern juridical or scientific formations and, as such, would provide a privileged access to the body politic or the machine polity. Just as the ‘carceral system’ combined in the prison ‘discourses and architectures, coercive regulations and scientific propositions, real social effects and invincible utopias, programmes for correcting delinquents and mechanisms that reinforce delinquency’ (p. 271) to stabilize modern jurisprudence, we might argue that the modern ‘experimental system’ is materialized in the laboratory to coordinate a complex ensemble of the resources and constraints that constitute modern science (Shapin, 1984, 1988). ⁴ Both capture the reality of a regime that mobilizes truth to attain power and deploys power to institute truth.

Shapin and Schaffer chose a canonical episode involving a figure long identified as ‘a founder of the experimental world in which scientists now live and operate’ (pp. 4–5) rather than a neglected ‘event’ that would disperse the myth (Canguilhem, 1997; Johnston, 1990). Nevertheless, they did not restore an unbroken continuity between Boyle’s laboratory and ours, but dismantled our naturalized myth about the logical relationship between experimental facts and mathematical representation. Hobbes’ ideal power–knowledge regime was plotted with mathematics in lieu of experiments, while Boyle’s relied on experiments without mathematical representation. Newton’s and our philosophical generalization that tightly binds experimental facts to mathematical theories did not happen at the birth of modern laboratory but much later in the 19th century through a complex cultural negotiation (Smith and Wise, 1989). What is found at the reputed ‘origin’ of modern physics is not the ‘inviolable identity’ (Foucault, 1984: 79) of a timeless experiment and a true mathematical representation but a ‘dispersion’ of what we recognize as the essential components of modern experimental science and ‘a space of multiple dissensions’ that destabilize our assumptions about what modern science is (Foucault, 1972: 72, 149–156).

Toward a material geography of the laboratory

Foucault’s (1980) focus on the local as the marginal highlighted the reality of domination by global knowledge–power complexes to suggest the fragility of such violence-ridden systems of knowledge. He hoped that his archeology and genealogy would work together to reactivate subjugated knowledges for the history of the present that would serve as an antidote against teleological and totalizing histories. ⁵ While his archeology carried structuralist resonances in the longue durée analysis of centuries and in the continental coverage of epistemes, ⁶ his genealogy would produce ‘fragments’ to expose the ‘infinitesimal mechanisms’ of power. Panoptism, ‘an ensemble of mechanisms’ that implemented power at the local level for an ‘integral surveillance’, had to be generalized to establish a ‘universal gaze’ that supports the state apparatus (pp. 71–72). Exactly how a regime of power–truth–knowledge stabilized at one locale could be extended over space and time remained necessarily vague, however, despite his painstaking description of how ‘the prison transformed the punitive procedure into a penitentiary technique’ and how ‘the carceral archipelago transported this technique from the penal institution to the entire social body’ (Foucault, 1973b: 298). Notwithstanding his pointed critique of the prevailing ‘descending analysis of power’ (e.g. Marxist), Foucault (1980) did not articulate how one should go about doing ‘an ascending analysis of power’ that would show how the micro-mechanisms of power have been integrated ‘by ever more general mechanisms and by forms of global domination’ (p. 99). We might consider his project as that of envisioning an alternative system of knowledge to the modern science–power complex, rather than explaining its historical genesis. How his genealogy might characterize the spatiotemporal integration that has constituted the modern episteme and the material empire of science remained unanswered, or perhaps unasked.

Shapin and Schaffer did not take it for granted that the power regime and the material discipline stabilized at one locale could be sustained over space and time. The laboratory as a culture-specific space–time complex (Shanahan, 2008) made it unlikely that it could be transported effortlessly to spread universal, timeless truth (Nagel, 1986). In other words, they raised a fundamental challenge to historians, philosophers, and sociologists of science: How does a scientific fact acquire its universal, timeless character (Jennings, 1988)? They criticized our ‘unreflective membership’ in modern scientific culture and sought to deconstruct ‘the member’s self-evident method’ by ‘playing the stranger’, or by adopting ‘a calculated and informed suspension of our taken-for-granted perceptions’ to discern alternative beliefs and practices (p. 6). While some internalists praised the book as a vigorous, original, and thought-provoking historical analysis that strengthened the importance of the Scientific Revolution (Cohen, 1987; Hankins, 1986), Leviathan and the Air-Pump challenged the self-evident nature of experimental truth and the transparency of instruments to expose the social conventions that constituted scientific facts and thereby modern science. If Shapin and Schaffer rode on the ethnographic turn that prioritized local histories and thick descriptions (Geertz, 1985; Rudwick, 1985), their purpose was not simply to write an exquisite microhistory for our antiquarian pleasure (Nietzsche, 1949), but to denaturalize the self-evident legitimacy of Western science by exposing the violence/discipline that went into its construction. Bruno Latour (1987) has pushed this reflexive point further in calling for a comparative anthropology of technoscience that would dismantle ‘the modern constitution’ (Latour, 1993). Whether we live in an amodern, postmodern, or post-postmodern world is not an idle speculation, but one that calls for conscious action to recast the place of science, humans, and nonhumans in it. If we have never been modern and science has never been universal, what kind of world are we trying to build with science?

If we can no longer assume that Western science has spread effortlessly to the rest of the globe simply because of its truth to nature, how do we account for its triumph? The growing research on the relationship between science and empire during the past decades has revolved around two models. Latour (1987, 1988) has stipulated a mechanism of information extraction at the ‘centers of calculation’ that operate the ‘cycles of accumulation’ to favor a differential growth of knowledge and power at the imperial centers, similar to Francis Bacon’s (2002) vision of a moral empire based on scientific/technological mastery in New Atlantis (White, 1958). The island of Bensalem possessed the languages, books, and affairs of foreign lands as if it were ‘a land of magicians, that sent forth spirits of the air into all parts, to bring them news and intelligence of other countries’ (p. 466). The island as the laboratory of the commonwealth required a spatial mastery and a material control to serve the imagined nation/empire. Londa Schiebinger’s (2007) notion of ‘green gold’ captures the importance of botany in the mercantilist economy that shaped the European empires. While she pays attention to biocontact zones, indigenous knowledges, their selective transfer to Europe (agnotology), and imperialist naming games, her vocabulary (e.g. bioprospecting) does not question the logic of the mercantilist economy, a system that took an enormous effort and violence to be instituted and maintained (Foucault, 1973b: 66–214; Schaffer, 2002; Vilches, 2010). Regardless of their intentions, that is, unsuspecting users of Latour’s and Schiebinger’s vocabulary have piled up case studies that testify to the practical necessity and utilitarian efficacy, if not the moral legitimacy, of European empires.

It is important to recover a geographical problematic in Leviathan and the Air-Pump, then, because it configured something like the provincial origin of European science (cf. Chakrabarty, 2000) and pointed to the discipline/violence required to make that science universal by delineating the technologies and skills that constituted the boundaries of an experimental community. In Shapin and Schaffer’s account, the material, social and literary technologies each functioned as an ‘objectifying resource’ – the machine for human perception, witnessing for the laboratory space, and experimental narrative for the larger public – in producing and naturalizing matters of fact. This process of objectification subjugated the machine and its technicians to the gentlemen, however, to make them invisible as the naturalized components of modern laboratory science. By tracking how the air-pump was disseminated, constructed, and modified during the first decade of its operation, therefore, they showed how troubled replication was and how limited its geographical displacement, to question the transparency of the air-pump and the invisibility of the technicians in producing matters of fact (Schaffer, 1992; Shapin, 1989). A European knowledge space of experimental philosophy depended on replicating the machines and the experiments in multiple contexts. Their full-bodied description of how experimental knowledge traveled in a piecemeal fashion over a limited distance, constantly subject to modifications, pointed to the need for a material geography of scientific knowledge. The spatial and temporal transmission of experimental knowledge was an emergent patchwork that depended on local actors, institutions, and polities and required constant recalibration and coordination (Schaffer, 1991), similar to the way in which Renaissance Portolan charts were made by calibrating real space–time explorations (Turnbull, 1996).

Foucault was of limited use in charting the historical geography of knowledge–power systems, although his spatializing gaze on the body as an emergent object of inquiry allowed medical geographers to probe the internal complexity of the body and the clinic to cultivate geographical perception and historical understanding of their modern normality (Philo, 1992, 2000). He did share the vocabulary and the problems of postcolonial geography in deconstructing the Eurocentric systems of knowledge (Crampton and Elden, 2007; Livingstone, 1995). Foucault (1980) hoped that his genealogy, analyzing the ‘tactics and strategies of power … deployed through implantations, distributions, demarcations, control of territories and organisations of domains … could well make up a sort of geopolitics’ (p. 77). Moving away from the representations to the technologies of geopolitical systems would require rematerializing our imperialist cartography, however (McEwan, 2003; Tolia-Kelly, 2013). While postcolonialism has successfully deconstructed the unity and centrality of Europe in global history, geographers have not seriously questioned the transhistorical objectivity of the maps that have provided the most powerful tool of imperial control at a distance. The postcolonial world still lives in the colonial space of representation, crystallized in the ‘objective’ maps produced by the European empires as ‘the essential preparatory step for possession, control, [and] mastery’ (Harley, 1988). Making knowledge universal required an imperial knowledge space (Turnbull, 1996).

In order to understand the emergence of a global knowledge–power complex, scientific practices need to be located, localized, and put in relation to each other spatially and historically. To this end, Renaissance geography/cosmography can offer us a heuristic counterview of the globe, since it comprised all physical, geographical, anthropological, economic, political, and religious knowledge until the Spanish Empire prioritized and institutionalized the mathematical grid in the Ptolemaic fashion. ⁷ The restoration laboratory was not just a point on the map, but a material place that organized social activity and generated a political imaginary to constitute a local knowledge–truth–power complex (Ophir and Shapin, 1991; Shapin, 1998; Withers, 2009). In other words, we need to recover the Renaissance experience of space as an emergent quality of exploration to build an alternative ontology and representation of space (Curry, 2005; Marston et al., 2005), perhaps akin to Leibniz’s relational space of active entities in contrast to Newton’s absolute space that contains bodies. Such a cosmographic (rather than cartographic) planetary consciousness can envision an alternative global polity that caters to local differences while ensuring global cooperation (Robinson, 2003). ⁸