In recent posts, I’ve laid out the [DC:REL] crossing. This is the pernicious amalgamation to which religion has been subjected in Modernity and that has caused its rationality to be obscured and abused.
Having provided this diagnosis—and with the bitter taste of its artificial expression in our mouths, in the form of fundamentalist violence—I think it’s time for us to move on, to the promised land, to religion itself.
So let’s talk about religion as a mode of existence: [REL].
But let’s pause again… For the first step in detecting [REL] will require the identification of (what I like to call) an ‘empirical site’. To explain why this is necessary, it will first be useful to recap the definition of a mode of existence. For Latour, modes of existence are revealed where experience registers a discordance with the way in which it has been officially described, designated and contained within the Modern constitution (the famous ‘practice’ versus ‘theory’ divide), that is, at moments where it finds itself ‘corseted by too narrow a set of legitimate agencies’ (Latour, Waking up from ‘Conjecture’ as well as from ‘Dream’: A Presentation of AIME, 2013, p.3). Modes of existence can be dis-amalgamated at points of discordance like these.
This empirical methodology is constantly foregrounded in the Inquiry: Latour whimsically describes it as being akin to the gentle tapping of Nietzsche’s hammer or the delicate wielding of Occam’s razor upon the surface of Modern experience, sounding it out in such a way as to determine where its claims to rationality ring hollow. Where it does sound hollow, the value of a mode of existence can be extracted.
The notion of an empirical site has been highly contested in responses to Latour’s Inquiry. I would refer you in particular to Terence Blake’s blog (I can’t recommend Terence’s trenchant analysis highly enough, even if we have different evaluations of Latour’s writing on religion), see here or any number of related links you will find from there. I’m not going to wade that controversy here: my aim is more to outline something of the structure of these empirical sites—and why on earth Latour deems them worth the trouble (for indeed he does deem them worth the trouble, in fact, they are intrinsic to the methodology).
The notion of an empirical site should be understood in a twofold way. First, it is a concrete, geographical location, one that can be visited and documented via processes of thick description. But second, it represents a kind of abstract, quasi-laboratory environment, one in which the thick descriptions provided by respondents in situ are put under the microscope and examined for their ontological consistency. This twofold function mirrors the job specification required of the co-enquêteur for the Inquiry, which should be understood as being a synthesis of the roles of ‘anthropologist/ ethnographer’ and ‘philosopher’.
Every single mode of existence Latour identifies is activated at an empirical site. For example, the detection of [REF] was activated by fieldwork carried out in the laboratories of the Salk Institute in California and the detection of [LAW] was activated by participatory observations carried out over a two-year period spent in the Conseil d’État in Paris.
In fact, in an early work Latour recounts a strange story that took place in his own life and that can be read as a proto-‘origin’ or proto-‘legitimation’ narrative for his entire method of enquiry, perhaps akin to Descartes’ poêle narrative (Latour, The Pasteurisation of France, 1993, pp.174–176). This story carefully incorporates the twofold structure of an empirical site described above and therefore seems to represent a sort of ‘macro’ site that pre-emptively orientates and envelopes the whole method of enquiry that will be pursued in the corpus that follows (Graham Harman is therefore right to describe this narrative as the ‘primal scene of Latourian philosophy’).
The movement by which modes of existence are extracted from empirical sites is crucial to grasp. Latour does not propose that an empirical site will straightforwardly yield a mode of existence to the enquêteur: after all, an empirical site still resides within Modernity and will correspondingly be subject to its institutional purification. Thus, it would be a mistake to suppose that an empirical site can be mined as a ‘representative cross-section’ (échantillon représentatif) of a mode of existence, in such a way that it might be scaled up and generalized at any point (Latour, Pour une ethnographie des modernes, 2008, p.4). This would be a confusion (and it is often confused in this secondary literature). No: an empirical site must be understood as furnishing a ‘contrast’ (contraste) (Latour, Pour une ethnographie des modernes, 2008, p.4). This term is to be understood in the mundane sense of holding up a sample or swatch to the light in order to ascertain where the fabric is lighter or stronger: empirical sites are not themselves modes that can be compared against each other, rather, an empirical site contains in nuce the value that inheres to a particular mode of existence. Thus, what matters is not so much the accumulation of examples (how many case studies can be secured and documented from this empirical site so as to validate a theory about its operations?) as an intense form of immersion (how does the actual performance of entities within this empirical site reveal a value different to the one instituted by Modernity?). This generates the necessary ‘contrast’: ‘a contrast is obtained because one has progressed far in an example, and not because one has generalized out from a cross-section’ (un contraste est obtenu parce que vous avez descendu loin dans un example, et non parce que vous avez généralisé à partir d’un échantillon’) (Latour, Pour une ethnographie des modernes, 2008, p.4). A contrast enables an empirical site to be ‘shifted up’ to a mode of existence.
What, then, provides the empirical site for religion as a mode of existence? It turns out that the situation here is complex—and part of the question we will have to tackle is whether this complexity is productive (generative, fecund, exciting) or problematic (infective, self-contradictory, collapsing in on itself). For in the case of [REL] the twofold definition of an empirical site seems in danger of failing. What is the concrete, geographical location from which Latour proposes to extract the contrast? It is not immediately clear. One candidate is found in Rejoicing, where Latour recounts (without specific detail) his visit to an ancient church in Montcombroux-les-Mines in the Allier department of central France (Latour, Rejoicing, 2013, p.10). There are alternative candidates: for example, during the course of a private interview I have conducted with him (and hinted at here and there in his private writings), Latour has recounted private experiences of his own where it seems he first identified some of the rationalization and derationalization procedures characteristic of the Religion of the Moderns. The work of Charles Péguy would be another. In none of these cases, however, does Latour seem to delineate these as empirical sites with the same rigour as he does elsewhere.
The reason for this soon becomes clear. In the case of [REL], rather than a concrete, geographical location, Latour will specify an abstract, non-specific human experience to serve as empirical site. This turns out to be the experience of words of love that are shared between two people. Latour gives a different nomenclature to this situation in different texts, but here, for the sake of convenience, it will be referred to as ‘amatory speech’.
Before proceeding, it is important not to pass by too quickly the idiosyncrasy of this situation within the framework of Latour’s thought. Why is it the case that [REL], alone among the fifteen modes, should be characterised by such an abstract—rather than concrete, geographical—empirical site? At various points Latour attempts to neutralise this as a potential problem or abnormality within his own system. He suggests, for example, that amatory speech can be taken as concrete insofar as it is a universal human situation, understood by all people at all times and places in the same way (Latour, An Inquiry into Modes of Existence, 2013, p.307). And yet, having said this, he then seems to qualify this assertion by suggesting that amatory speech represents merely a ‘prefiguring’ or ‘scale model’, thereby somewhat undermining the claim to universality he had previously made for it (Latour, Rejoicing, 2013, p.118; also cf. p.48).
It seems to be the case, then, that a certain ambiguity inheres in the empirical methodology that pertains to [REL]. This tension, if it cannot be resolved, would bear serious consequences, for if it could be shown that [REL] is derived as empirical correlate of a prior hermeneutical decision then its veridiction would be compromised.
That is the question that we will have to face as we begin to delve into [REL] more and more in the posts to follow.
 For references to Nietzsche’s hammer; cf. Gifford Lecture 4 (2013), pp.75, 78. For references to Occam’s razor; cf. Waking up from ‘Conjecture’ as well as from ‘Dream’: A Presentation of AIME (2013), p.1.
 ‘Amatory speech’ as empirical site does not imply a reduction of [REL] into merely a discursive phenomenon. The words of love that are communicated in amatory speech are understood by Latour as ‘entities’ performing according to the logistics of [NET:PRE], and thus as equivalent to any other sort of material entity.
 Which does seem to be acknowledged by the fictional ‘enquirer’ that is used as a mouthpiece in the Inquiry, who is described as being ‘[…] a little annoyed with herself for having to limit [religion as a mode of existence] for the moment to the intimacy of love crises’, in Latour, An Inquiry into Modes of Existence, 2013, p.305.