Category Archives: Latour

Latour on factishes and belief

Before moving on to work on Spinoza and the concept of aberrant monism, I want to add one more post on Latour. I hope that between this and previous posts there may emerge a relatively coherent picture of my reading of Latour. I also hope to indicate how Latour’s thought can become, and ought to become, an effective tool in countering a number of the presuppositions of contemporary neoliberal politics, or what Mark Fisher aptly calls capitalist realism. As usual, feel free to point out the errors of my ways.

As contradictory as it may seem, Latour argues that ‘construction’ and ‘autonomous reality’ are to be understood as synonymous. Among the neologisms Latour uses to elaborate this point is ‘factish,’ being a combination of fact and fetish. A fact, traditionally understood, is autonomous and unconstructed. When Pasteur discovered the role microorganisms play in the process of fermentation he simply, on this view, came to recognize an autonomous fact that was already there and independent of the historical events involved in Pasteur’s efforts to locate these microorganisms. A fetish, by contrast, involves the projection of beliefs upon a mute, passive object. In both cases, according to Latour, what is maintained is the subject-object dichotomy. In the case of facts, objects are ultimately responsible for the success of Pasteur’s experiments; and in the case of fetishes, subjects are the ones responsible for projecting beliefs onto objects. A factish, for Latour, is a type of action that does ‘not fall into the comminatory choice between fact and belief.’ (Pandora’s Hope, 306). Rather, a factish entails events; or, as Latour puts it, ‘I never act; I am always slightly surprised by what I do. That which acts through me is also surprised by what I do, by the chance to mutate, to change, and to bifurcate…’ (ibid. 281). In a nod to Deleuze, Latour claims that factishes are ‘rhizomelike,’ or ‘one should always be aware of factishes…[because] their consequences are unforeseen, the moral order fragile, the social one unstable.’ (ibid. 288).

A factish is thus neither an independent reality that comes to be discovered after a successful scientific experiment, nor is it merely the projection of human beliefs onto an inert object. A factish involves both human and nonhuman actors, and scientific experiments, as events, involve relations between actors that return more than any of the actors contributed singly (neoliberal understandings of the market ignore this fact, but the recurrence of crises, as David Harvey has argued, justifies Latour’s point). It is for this reason that Latour claims ‘an experiment is an event which offers slightly more than its inputs…no one, and nothing at all, is in command, not even an anonymous field of force.’ (ibid. 298)(Nor, for Latour, would an anonymous, impersonal market be in command). Latour acknowledges the influence of Whitehead when he uses the term event, most notably the use Whitehead makes of the term to replace ‘the notion of discovery’ and its attendant assumption concerning the ahistorical nature of objects and the historicity of human endeavors. When Latour defines an experiment as an event, therefore, he intends precisely to argue that this ‘event has consequences for the historicity of all the ingredients, including nonhumans, that are the circumstances of that experiment.’ (ibid. 306).

Despite all the apparent differences between modernism and postmodernism, Latour argues that they have each ‘left belief, the untouchable center of their courageous enterprises, untouched.’ (ibid. 275). In particular, the modernists, on Latour’s reading, felt the task of philosophy and science was to track down which of our beliefs are justified true beliefs. To this end, science comes in armed with facts to hammer away at any beliefs that are not in line with the facts. With postmodernism, on the other hand, science itself comes to be seen as nothing but beliefs that construct a reality, or ‘construction and reality are the same thing; everything is just so much illusion, storytelling, and make believe…’ (ibid.). And what we have then, ‘when science itself is transformed into a belief’ is ‘postmodern virtuality—the nadir, the absolute zero of politics, aesthetics, and metaphysics.’ ‘Virtuality,’ in short, is for Latour ‘what everything else turns into when belief in belief has run amok.’ (ibid. 287). When Latour argues that it is because factishes are constructed that they are so very real, therefore, he is saying something quite different from the postmodernist who claims that ‘construction and reality are the same thing.’ The difference has to do with who or what is acting. In the case of factishes, there is a ‘rhizomelike’ network of both human and nonhuman actors, and no one actor is in command, ‘not even an anonymous field of force.’ For the postmoderns, it is the power of belief that is in command, or the virtuality of the One as Badiou interprets Deleuze, and the autonomous nature of reality is nothing but a mode of belief or virtuality.

If Badiou’s critique of Deleuze is correct (as an earlier post indicated it was not), then it would seem that Latour would echo Badiou’s criticisms, at least on this point. As we unpack precisely what Latour is arguing for when he claims that it is ‘because it [a factish] is constructed that it is so very real,’ we will find that it bears much in common with Deleuze’s project, as Latour himself recognizes, and thus Badiou’s criticisms are directed at a ‘postmodern’ Deleuze that never was. Key to understanding how ‘construction’ and ‘autonomous reality’ are synonymous is what Latour calls ‘historical realism,’ whereby the reality of what is is inseparable from processes that increase or diminish the number of associations that are accumulated over time. It is not all or nothing regarding the existence of entities, but we ought rather to speak of ‘relative existence.’ As Latour puts it, ‘An entity gains in reality if it is associated with many others that are viewed as collaborating with it. It loses in reality if, on the contrary, it has to shed associates or collaborators (humans and nonhumans).’ (ibid. 257). For Deleuze the heterogeneous network of human and nonhuman associations that constitutes the relative existence of an entity is just what he calls a multiplicity, or what I have called historical ontology, the double articulation and movement that is inseparable from the reality of entities. Moreover, we could say that the reality of entities is a reality-effect of historical ontology, though an autonomous effect whereby historical ontology is a quasi-cause (as Deleuze understands this term). Using Deleuze’s terminology, the reality of an entity is inseparable from a double articulation, with the first articulation drawing a number of associations and links between human and nonhuman elements into a plane of consistency, and the second articulation actualizing this plane of consistency as a real, autonomous entity. As Latour and Woolgar state it in Laboratory Life, ‘“reality” cannot be used to explain why a statement becomes a fact, since it is only after it has become a fact that the effect of reality is obtained.’ (p. 180). For Deleuze the multiplicity is the reality of the virtual that is the quasi-cause inseparable from the reality of entities that are its autonomous effects. A further and related concern of Deleuze is what he calls counter-actualization, or a problematizing history as I call it, whereby the virtual as quasi-cause becomes tapped, thereby problematizing the actual so as to allow for its possible transformation. Latour’s use of history in science studies shares a similar concern. In particular, by problematizing traditional understandings of science and the presuppositions it entails concerning the relation between beliefs and things, it could be argued that Latour is attempting to problematize and hence move beyond belief. Rather than attempt to justify beliefs through scientific facts or unmask beliefs as mere fancy and fetish, Latour sees ‘the role of the intellectual’ as the task of ‘protect[ing] the diversity of ontological status against the threat of its transformation into facts and fetishes, beliefs and things.’ (Pandora’s Hope, 291). In short, Latour, as with Deleuze, sees the ‘role of the intellectual’ as one of affirming multiplicity in order to counter the identification of multiplicity with the one or multiple of beliefs and things. And it is with this move that Latour’s thought has begun to engage more actively and critically with a number of the presuppositions of contemporary neoliberal politics, as I’ve tried to indicate briefly here and in previous posts.


Events and Objects (à la Latour)

In We Have Never Been Modern, Latour adopts a cartographic metaphor to explain the stabilization of the Nature-Society duality from unstable events, or the trajectory from A” to D” marks, as Latour puts it, ‘the gradient that registers variations in the stability of entities from event to essence.’ (p. 85). Adopting Spinozist language, Latour will later clarify his ontology; namely:

…the immanence of naturing-natures [Nature] and collectives [Society] corresponds to a single region: that of the instability of events, that of the work of mediation…If we add to the official, stable version of the Constitution [i.e., the Nature-Society divide or gap] its unofficial, ‘hot’ or unstable version, the middle is what fills up, on the contrary, and the extremities [the Nature and Society poles] are emptied. (ibid. 86-7).

A very similar point is made in Latour’s early work, Laboratory Life, when discussing the status of scientific objects, in particular TRF (thyrotrophin releasing factor), he argues that ‘the solidity of this object, which prevented it from becoming subjective or artefactual, was constituted by the steady accumulation of techniques.’ (127). A consistent theme for Latour, then, though he did not always use the same terms to detail it, is that events and objects are not to be confused with one another even though they are not fundamentally distinct, much less different in kind. Rather, objects are stabilized events; or, adopting Latour’s own metaphor, objects are ‘the cooled down continents of plate tectonics.’ (Never Been Modern, 86). Objects are thus inseparable from their unstable networks, and even though an object may be particularly stable and even lionized as a textbook fact, it may lose allies to another object that, through a ‘steady accumulation of techniques’ and alliances, displaces it. Latour gives the example of how prior to Watson chemists preferred, and textbooks stated as an established fact, that the four DNA bases were in the enol form which subsequently made it more difficult for Watson to cast doubt upon it and put forth the case that it was in the keto form instead. (Lab Life, 243). A Latourian ontology is monistic, therefore, in that there is nothing but objects and events – the middle is filled and the poles are empty as he put it – but this is not a stable monism of autonomous objects and lawful events; rather, it is an aberrant monism that continually moves between stable natures and collectives and unstable, aberrant events. What implications does this have for placing Latour within the OOO camp? It seems to place him on the margins, though I’m not committed to this.


Latour, Whitehead, and Chaosmos

Where Latour and Whitehead most clearly converge is in their emphasis upon events, and in particular with their understanding of events as always exceeding themselves. Latour has developed this line of thinking in numerous works, but most notably in We Have Never Been Modern, and he among others (especially Isabelle Stengers) have frequently recognized the significance of Whitehead on this point. For Latour as for Whitehead (and for Davidson [see earlier posts]) there is a single ontology of events, and these events function as mediators that continually work, rework, and transform other events. Over time these events come to be stabilized as either natural events and phenomena or as subjective and cultural artifacts. The modern view Latour contests, by contrast, views events as intermediaries from the start that already embody their subjective or objective essence, an essence that will be revealed. In the first case, events are historical and anti-teleological in that they reflect the contingencies of their relationships to other events that are taken up over time though without presupposing whether such collectives (as Latour calls them) are guaranteed of success. In the second case, events are ahistorical and teleological. The essence is already there and whatever contingencies occur are accidental to the nature of the event, a nature that is predestined to be revealed at the end of the day.

We can better understand Whitehead’s position if we look at his critique of Hume. For Whitehead Hume introduces the world ‘as a secondary conjecture’ that is constituted on the basis of a multiplicity of discrete impressions and ideas. Whitehead, by contrast, argues that discrimination itself is exercised on the basis of an ‘experienced world,’ an experience that ‘starts,’ Whitehead claims, ‘with a sense of power, and proceeds to the discrimination of individualities and their qualities.’ (Modes of Thought [1968], p. 119). In other words, we do not begin with ‘the many data’ and then construct an experiential unity of the world; or, as Davidson understands it, we do not begin with a multiplicity of empirical data that is then taken up and organized by a conceptual scheme (the third dogma of empiricism). We begin, Whitehead argues, with an experiential unity and power that Whitehead characterizes as the ‘compulsion of composition,’ or the process whereby a felt unity discerns and discriminates (or prehends to use Whitehead’s term) novelties that perpetuate the process of composition. For Davidson, similarly, we begin with a general agreement that serves to make intelligible the differences and discriminations about which we subsequently disagree and argue. It is therefore not that which is discriminated that is most real – in contrast to Hume – nor is it a completed, self-sustaining composition – in contrast to the rationalists – but it is instead the compulsion of composition itself which is most real for Whitehead. Yet despite his critique of Hume, Whitehead nonetheless claims Hume is indispensable:

This conclusion that pure sense perception does not provide the data for its own interpretation was the greatest discovery embodied in Hume’s philosophy. This discovery is the reason why Hume’s Treatise will remain as the irrefutable basis for all subsequent philosophic thought. (Ibid, p. 133).

In other words, the data of sense perception do not, by themselves, provide the means whereby they become absorbed by a unifying interpretation. Although on his reading of Hume the sense data is indeed what is taken up and organized by an interpretive process (hence Hume is committed to the third dogma), for Whitehead there is no sense data or data of any kind that exists independently of being taken up by an interpretive process. Whitehead avoids the scheme-content dualism (third dogma) in much the same way Davidson does—namely, by calling for an ontological monism, in this case a monism of actual entities: ‘apart from the things that are actual, there is nothing.’ (Process and Reality, p. 53). And the actual entities that constitute actuality are to be ‘conceived,’ according to Whitehead, as an act of experience arising out of data. It is a process of ‘feeling’ the many data, so as to absorb them into the unity of one individual satisfaction.’ But this data is nothing less than other actual entities, and thus each actual entity is an event, a process, whereby it is simultaneously a subject that prehends other actual entities and ‘absorb[s] them into the unity of one individual satisfaction’ and it is in turn a unity that can be taken up within the prehensions of other actual entities, and so on—actual entities are thus both subject and superject as Whitehead puts it.

By adopting a monistic ontology of events, it is no longer a question for Whitehead of having to construct, amidst a multiplicity of already identified and identifiable entities, the necessary relation between them such as cause and effect, a subject’s belief about the world and the way the world is, etc. Rather, for Whitehead an event develops amidst a chaos of actual entities and through a process of composition and screening there is the coming to be of a stable world or cosmos. Consequently, it is the very identifiable, determinate and stable nature of the entities that are which marks for Whitehead the end of actual entities, the end of the screening process. Only as determinate facts after the end of process do entities then embody certain relationships – such as cause and effect, subjective or objective, etc. The screening process thus does not exclude or repress any determinate and identifiable entity. Deleuze and Foucault both echo Whitehead on this point, and this explains why they reject Freud’s view of the unconscious. (Difference and Repetition [1995], pp. 106-7). The screening process therefore does not function for Deleuze, and for Whitehead as Deleuze reads him, as the guard Freud discusses in the metaphor he uses to explain repression. In the metaphor there are two rooms. One room contains entities that we are conscious of. Some of these entities are the focus of our conscious attention while the rest are part of the background of our conscious awareness. The second room consists of entities that are in our unconscious. Between the two rooms is a door and a watchman who monitors who can or cannot move from the unconscious to the conscious, or what, conversely, of our conscious life gets put into the unconscious. The problem with this theory for Deleuze is that it results, at worst, in a form of ontological dualism, or at best in a continued adherence to the third dogma of empiricism. If the relationship between events and chaos is taken to be one where a chaotic realm of entities is forged into a stable cosmos, then we would again be back to the view whereby a conceptual scheme is inseparable from a screening that excludes those entities that are from the start outside all conceptual schemes – in short, we would be back to the third dogma. For Deleuze, by contrast, the unconscious is not distinct from the conscious, the chaotic is not a separate and distinct realm from cosmos. By an ‘unconscious in finite understanding,’ Deleuze means that there is ‘something that cannot be thought in finite thought…a nonself in the finite self…’ (The Fold, p. 89). In short, there are nothing but actual entities; or, as Whitehead claims, ‘apart from things that are actual [i.e., actual entities], there is nothing.’ The screening that stabilizes events is nothing less than the self-organization of actual entities themselves and reflects the fact that events, as processes and becomings, already exceed themselves and are assured of becoming other.


Marx or Tarde

In a post over at Bogost’s blog there’s an interesting discussion of Marx. I think it is correct to say one is not a Marxist if by that one means that Latour is not a Marxist. Latour is explicit on this point in his recent book, The Science of Passionate Interests, where he wonders how the 20th century might have unfolded had Tarde’s approach to understanding capitalism been more influential than Marx’s. Latour’s critique of Marx is much the same as his critique of Durkheim (you could substitute Durkheim for Marx in the previous sentence). Rather than presuppose the existence of class and society, Tarde examines the myriad ways in which society is composed. Latour follows a similar approach, of course, and in an essay he wrote with Shirley Strum, ‘redefining the social,’ he explicitly claims that society is not a given but needs to be composed, and composed by way of things – i.e., our human/nonhuman interactions. I develop this argument extensively in Deleuze’s Hume. Thanks to Robert and Ian for bringing my attention to Latour’s unpublished “Compositionist Manifesto.”


Is Deleuze a Speculative Realist?

At first it might seem he is. If Bruno Latour is on the right track with respect to speculative realism, as Graham Harman and others would argue, then it might seem that Deleuze is on the right track as well for there are a number of areas where their philosophies converge in significant ways – especially concerning events, multiplicity, and their embrace of an ontological monism. I cover much of this in Deleuze’s Hume. It would also seem that Deleuze would not be a “hyper-incommensurable” postmodern philosopher as Latour discusses this in We Have Never Been Modern. Not only does Lyotard, for example, continue to embrace the incommensurability between humans and nonhumans, but will go even further and claim that ‘there is nothing human about scientific expansion,’ thus radicalizing the incommensurability (hence the ‘hyper-‘). Deleuze, by contrast, moves with ease in discussing human and nonhuman assemblages. The frequently used example of an assemblage – man-horse-stirrup – is a case and point of Deleuze (and Deleuze and Guattari’s) readiness to at the very least blur if not eliminate the incommensurability between humans and nonhumans.

Then there is Quentin Meillassoux’s critique of the correlationists. One of the central planks of the SR platform is the critique of correlationism. Kant is frequently singled out as the subtle grandmaster of correlationism (the final chapter of After Finitude sets out to undermine [and correctly so I might add] Kant’s Copernican revolution). But Meillassoux doesn’t simply have his sights set on Kantians; rather, he sees much if not all of the post-Kantian philosophical tradition as beholden to certain correlationist assumptions, or to what Harman calls a ‘philosophy of access.’ Most notably, the correlationist philosopher thinks that our only access to objects is through thought. Thus, we cannot think the thing in itself but only as given to thought, as a correlate of thought. As Meillassoux points out, however, a correlationist is not necessarily tied to a subject-object metaphysics, to a hypostasized subject and object in the manner of Descartes; rather, what is central to correlationism, at least since Kant, is ‘not a metaphysics,’ but rather ‘it invokes correlation to curb every hypostatization, every substantialization of an object of knowledge which would turn the latter into a being existing in and of itself.’ (After Finitude, p. 11). Correlationists, in short, cannot think an object as it is in itself and correlationism assures the impossibility of ever thinking an object in itself.

With Deleuze, for Meillassoux, we have a classic example of an attempt to ‘curb every hypostatization,’ and moreover we have in Deleuze and Nietzsche ‘the vitalist hypostatization of the correlation’ as an integral aspect of their critique of metaphysics (ibid. p. 37). Without addressing the fact that Deleuze never saw himself as part of the ‘critique of metaphysics’ tradition, the question remains: is Deleuze hypostatizing the correlation with his notion of life and his emphasis upon process and becoming? In other words, if there are for Deleuze no objects, if objects are merely abstractions of a flux, much like Bergson’s snapshot photographs were abstractions of duration (duree), then objects would indeed simply be correlates and abstractions of becoming. Deleuze would thus be a strong correlationist, as Meillassoux argues, or a hyper-correlationist as Latour might argue. As for Kant it is impossible to think the object in itself but only as a phenomenal correlate of thought, is Deleuze a strong correlationist who, as Meillassoux argues, that ‘it is unthinkable that the unthinkable is impossible’? (ibid. 41). If there remains something that is unthinkable, namely becoming, and if it is indeed unthinkable that this unthinkable is impossible, then Deleuze would most definitely not be a speculative realist since a central task of SR is to think objects in themselves.

But what does it mean to think an object in itself, or as Meillassoux puts the problem, an object that is anterior to givenness itself? Put simply, it is to think the absolute, to think that which is not limited by being given to a consciousness, to a historical situation, discourse, etc., but to think the absolute in itself. This absolute, however, is not to be an absolute becoming, life, or will to power, for then we would be back in correlationism. Rather, the absolute, for Meillassoux, is contingency itself, or, as he puts it: ‘The absolute is the absolute impossibility of a necessary being.’ (Ibid. 60). It might seem that Deleuze would agree on this point, but it is precisely here where Deleuze would stumble, for in absolutizing becoming he ultimately calls upon a necessary being, a contradictory, paradoxical being. Meillassoux is quite clear on this point (as he is with most of the points he makes): ‘the utterly Immutable instance against which even the omnipotence of contingency would come to grief, would be a contradictory entity. And this for the precise reason that such an entity could never become other than it is because there would be no alterity for it in which to become.’ (ibid. 69). It would already include its contradictory other and thus such an entity would be a necessary being and hence undermine contingency itself. It is for this reason that to think the absolute one must not think it as becoming, if for becoming ‘things must be this, then other than this; they are, then they are not.’ (ibid. 70). ‘The only possibility of introducing difference into being, and thereby a conceivable becoming, would be by no longer allowing oneself the right to make contradictory statements about an entity.’ (ibid. 71). In the end, it is only through mathematics that one can think the absolute as the contingent without contradiction, and philosophers of becoming such as Deleuze, Bergson, and Nietzsche continue to affirm the right to utter contradictions, much as did their intellectual progenitor Heraclitus.

Deleuze’s philosophy, however, is not to be confused with Bergson’s and Nietzsche’s, despite the influence of the latter two on Deleuze’s own thought, and for Deleuze to think difference in-itself, as he claims is the central task of his philosophy in the early pages of Difference and Repetition, is not to think or to affirm a contradictory entity. As alluded to in my earlier post on Badiou and Spinoza, Deleuze rejects the idea that there is an identifiable difference, much less an identifiable contradiction, between two entities that it is the task of philosophy to think. This is not what it means to think difference in itself according to Deleuze. At the same time, identity for Deleuze is not merely a correlate of difference. Deleuze, however, does speak of the impossibility of thought, of an unconscious that is understood to be ‘something that cannot be thought in finite thought.’ (Fold, p. 89), or he will write in Cinema 2, in reference to Artaud and Blanchot, that “what forces us to think is ‘the inpower [impouvoir] of thought’, the figure of nothingness, the inexistence of a whole which could be thought.” (C2, 162). This impossibility and unconscious that thought itself cannot think but forces thought is not a necessary being (e.g., duration, becoming, will to power, a life etc.) relative to which what can be thought would merely be correlates of this necessary being. To the contrary, and much in line with Meillassoux, that which ‘cannot be thought in finite thought’ is ‘the absolute impossibility of a necessary being’, to quote Meillassoux again. It is ‘the inexistence of a whole which could be thought,’ or as Meillassoux understands it, adopting Cantor’s definition of transfinite numbers, ‘the (quantifiable) totality of the thinkable is unthinkable.’ (AF, 104). And the unthinkable nature of the totality is key to avoiding correlationism, for it is mathematics, especially Cantorian set theory, that is able to theorize the non-totalizable, the ‘non-All’ – hence Meillassoux’s conclusion that ‘what is mathematizable cannot be reduced to a correlate of thought.’ (ibid. 117). To the extent that Deleuze too theorizes the ‘inexistence of the whole,’ what Deleuze and Guattari will also refer to as the ‘nondenumerable,’ then it would seem that we would be too hasty to exclude Deleuze from the speculative realist camp. Moreover, as was argued in the Spinoza post, the non-denumerable, the ‘inexistence of a whole’ (i.e., or the attributes as discussed there) is not separable and distinct from the denumerable (the modes) and that which is thought. There is nothing but objects and events, both human and nonhuman, and there is no incommensurability between them nor are they totalizable in a way that would return us to claiming that objects and events are correlates of a necessary being. Deleuze is not a correlationist. But is he a realist? That will have to wait for another post.