In Defense of Latour (and his neo-Humean ways)

As most know, especially those who would be reading this post in the first place, interest in Latour has increased dramatically. No longer the eccentric novelty of being an anthropologist working at the Salk Institute and carving out a niche that came to be known as science studies (which in turn is to be contrasted with the sociology of scientific knowledge [SSK] school in Edinburgh), Latour has more recently come to be a central figure in a philosophical movement on the ascendancy – OOO. Graham Harman is quite forthright in his praise: Latour is for him ‘the closest figure [he] can think of to the ideal object-oriented hero.’ This is not to say that Latour’s thought fits seamlessly into the object-oriented ontology. It doesn’t, as I alluded to in an earlier post. In fact, there are a handful of criticisms that Harman and others make of Latour that I want to address briefly here, and will probably tackle at greater length in another context. Much of the thrust of the criticisms is diminished if one understands the role Latour’s ontology plays both in setting out a study of science that does not reduce it to being a branch of social constructivism, as Latour believes the SSK school does; and if one unpacks his politics as a compositionalism (which I’ll have to detail in a later post to keep this relatively short). But first the criticisms. The first two are from Harman and the third one will find here and here.

  1. Latour reduces objects to their relations.
  2. Latour begins with a symmetrical relationship between humans and nonhumans but ends with an asymmetrical relationship that gives power to humans and not objects.
  3. In doing away with the distinction between might and right Latour ultimately reduces normative force to causal force.

As for the first it does appear that Latour is guilty as charged. But why is this a problem? Primarily this fails to give to objects their proper autonomy. Only if an object is something more than its relations (of translation, causation, transformation, etc.) can we account for the becoming of this object. Harman gives the example of Heidegger’s Being and Time. If new interpreters of this text were to come on the scene, ‘what they would be interpreting,’ Harman argues, ‘is Being and Time itself, not the sum total of other interpretation’; that is, not the sum total of Being and Time’s relations. Harman thus concludes that ‘an actor is not identical with whatever it modifies, transforms, perturbs, or creates, but always remains underdetermined by those effects.’ (Prince of Networks, 186-7). There’s a few points to make concerning this conclusion. First, I agree completely that an object is underdetermined by its effects, and so does Latour on my reading. It was precisely on this issue where Latour broke ranks with the SSK school, and David Bloor in particular. On Bloor’s account a scientific theory is underdetermined by the observed empirical relations and hence something else needs to be added to the mix in order to account for why one theory is accepted over another when the facts themselves don’t mandate the one over the other. Bloor argues that it is the social, or social relations, that accounts for the eventual acceptance of one theory over another. Latour rejects this solution just as he rejects the naïve scientific realist who continues to believe that it is the object itself that guarantees the eventual acceptance or rejection of a theory (I am not equating OOO with naïve scientific realism). Nonetheless, Latour does accept the underdetermining of theories by facts, a claim one can find in Whitehead, and Whitehead himself claims that this is the indispensable legacy of Hume. As Whitehead puts it:

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. (Modes of Thought, p. 133).

So why is it not the object itself, not in the naïve realist sense, but in the OOO sense, the object as withdrawn, that accounts for the eventual interpretations and relations? I think for Latour the answer again stems from Hume, and in particular from Hume’s skepticism concerning causation. But really it is a skepticism concerning Descartes’ argument in the second Meditation. When Harman argues that ‘an actor is not identical with what it modifies,’ we’re back with Descartes’ claim that the cogito is not identical with its thoughts. Nietzsche found this argument unconvincing. As Nietzsche argues,

Modern philosophy, as epistemological skepticism, is secretly or openly Anti-Christian, although (for keener ears, be it said) by no means anti-religious. Formerly, in effect, one believed in “the soul” as one believed in grammar and the grammatical subject: one said, “I” is the condition, “think” is the predicate and its conditioned—to think is an activity for which one must suppose a subject as cause. Beyond Good and Evil

Then there is this quote (from Will to Power #484):

“Something is thought, therefore there is something that thinks”: this is what Descartes’ argument amounts to. But this is tantamount to considering our belief in the notion “substance” as an a priori truth: –that there must be something “that thinks” when we think, I merely a formulation of a grammatical custom which sets an agent to every action. In short, a metaphysico-logical postulate is already put forward here and it is not merely an ascertainment of fact…On Descartes’ lines nothing absolutely certain is attained, but only the fact of a very powerful faith.

So how then do objects come to be autonomous while at the same time not being independent of their relations? To ask it yet another way: how do we account for the underdetermination of an actor (actant or object) by its effects and relations without reducing it to the social in Bloor’s sense or an object in Descartes’ sense. This is where I think Latour adopts Whitehead’s and Deleuze’s metaphysics of the event. Put briefly, and absurdly simplifying, Latour follows Hume in claiming that there are processes inseparable from determinate, known causation but which are nonetheless irreducible to it. First there are habits, then there is knowledge. Latour will thus call for autonomous objects, objects that are irreducible to their relations—since 1864, for example, there have always been microorganisms, microorganisms independent of Pasteur’s work—and yet this irreducibility, this something more or underdetermination is not itself a set of actualized, determinate relations. It is what Deleuze will call the virtual, or more precisely it is the metaphysics of the event one finds in Deleuze and Whitehead. Every actual entity, as process and event, exceeds every other actual entity, including God (See Whitehead, Process and Reality [260]: ‘…every actual entity also shares with God the characteristic of transcending all other actual entities, including God. The universe is thus a creative advance into novelty.’) It is this move, this metaphysics (which I admit needs further elaboration and which I’ve detailed elsewhere), that enables a Latourian account of objects that does reduce them to their actual, determinate relations, but in a way that is nonetheless underdetermined by these relations.

This brings me to the second criticism – Latour’s move from a symmetrical to an asymmetrical relationship between humans and nonhumans. Whereas Pasteur is accepted to have preceded the existence of microorganisms, the same is not true for the microorganisms themselves; and this move runs counter to a basic tenet of OOO that there is no asymmetry or hierarchical differences between objects, whether human or not. First, it’s not true that for Latour microorganisms went from being nothing to something at the hands of Pasteur who was something all along the way. What was crucial to Pasteur’s success was that he was able to translate his laboratory findings into the fields and farms, that he translate the relations and effects observed in the laboratory to the observations had by farmers working with cows, beer, etc. A similar process occurs with respect to subjects. Pasteur no doubt was transformed by the experience as he went from being an obscure scientist to a leading light in the French Academy. Ian Hacking has also shown how we can come to reconceptualize and rethink, or translate, ourselves and who we are as subjects. What was once a misbehaving child is now a child with a medical condition, ADHD. The symmetry between humans and nonhumans breaks, for Latour, when Pasteur is given credit by textbooks for bringing microorganisms into being from nonbeing. This results only at the end of the process of the stabilization of events (as discussed here), and hence it is to prioritize the discontinuities of ready-made science over the continuities of science in the making. Latour clearly finds the work of science to be science in the making and ready-made science is the death of scientific work, much as an actual entity, for Whitehead, ‘has perished when it is complete.’ (PR, 99). This is not to say that there are no differences between humans and nonhumans. Some things do come to be artifacts rather than facts, are reflections of subjective bias rather than objective fact. For Latour this is a consequence of the strength and reach of networks rather than having anything to do with an objective or subjective essence that comes to be revealed through empirical studies. And if Latour devotes more time to studying the human component of these relations, or if he does not, as some have criticized him, address object-object relations, I believe this follows from his political concerns, as well as his academic training as a sociologist, rather than from a flawed ontology–hence his admiration for Whitehead.

And this brings me to the final criticism—namely, that Latour, in eliding the distinction between might and right ultimately reduces normative force to causal force. In other words, a position, theory, or argument ultimately wins the day not because it satisfied standards of rational coherence, consistency, and logical rigor, but rather because it makes the biggest splash and produces the most ripples throughout the networks. This is again where Latour’s neo-Humean streak comes to the fore. Latour certainly does not dismiss normative force (at least not on my reading). As Latour discovered and detailed in Laboratory Life and even in his most recent book, The Making of Law, the power of rational (and not rhetorical) argument is one of the factors that is brought into play in attempting to establish a claim. An argument that adheres more rigorously to the normative standards of rationality is thus more likely, all other factors being equal, than one that does not. But following Hume, and Whitehead as we saw above, the facts underdetermine the rational arguments that can be made concerning them. To adopt a cliché, reasonable people can disagree even if they accept all the facts of the case. This is why other factors need to be brought into play, and they are numerous, as a reading of Laboratory Life in particular makes clear. One of the problems Latour has with neo-liberalism, as he makes quite clear in his recent book, The Science of Passionate Interests, is that it reduces reality to a too simplistic quantity—our individual desires and beliefs as rationally determined. It is not that rational-choice theory is wrong but far too limiting to understand the complexities that constitute our social and political worlds. Latour is thus loathe, and he admits to following Tarde on this point, to reduce our understanding of the social world either to a set of normative primitives or to a mere quantifiable summation of causal effects. Things are much more complex than this, and forever in process. Latour thus does not reduce normative force to causal force but allows for normative forces, among others, to account for how underdetermined facts come to acquire their status as accepted, unquestioned facts and truths. Latour is therefore a true successor to Hume and Whitehead.

5 thoughts on “In Defense of Latour (and his neo-Humean ways)

  1. Pingback: a quick response to Bell « Object-Oriented Philosophy

  2. Hi Jeffrey,

    I think you’ve perhaps missed the point of the critique of Latour on normativity. The point is not that Latour cannot allow accounts of rational (or otherwise norm governed) behaviour to play a role in the explanation of various social phenomena. The problem is that he tries to give an account of normative force in terms of the *effects* normative statuses produce. The law of non-contradiction has force neither because one is causally compelled to follow it, or because following it is likely to produce to desired outcomes. It’s force consists in the fact that it presents a *good reason* to act in a certain way, in this case, to divest oneself of incompatible commitments. Normative force and causal force are not two species of a single genus, what I elsewhere called ‘effective force’, or that which produces effects ( The sense in which they are both forms of force is a matter of analogy. The post above doesn’t seem to take this point on board:-

    “Latour thus does not reduce normative force to causal force but allows for normative forces, among others, to account for how underdetermined facts come to acquire their status as accepted, unquestioned facts and truths. Latour is therefore a true successor to Hume and Whitehead.”

    If you treat normative force only as something which produces an effect (like the acceptance of facts), then you aren’t treating it as normative force. You’re already dealing with the force of causes rather than the force of reasons.

    To underline the point about neoliberalism which follows from this. My criticism was not that Latour is left explaining the social using something like rational choice theory, but that he is left to work out how he should act using something like rational choice theory. Of course he can hold that people generally don’t behave like ideal preference maximising machines (and he’s right), but he has no means to say that we shouldn’t do so. Indeed, it seems to be the only source of practical maxims he can provide.

    • Hi Pete,
      I do believe that Latour would have responded to the criticism regarding normative force roughly as I present it, but this follows from Latour’s more general effort to move beyond critique and Kant’s Copernican Revolution (see especially his essay, “Why has critique run out of steam?”). As Latour sees it, a deontological critique simply reinforces the modernist divide between a mute, ahistorical nature and a historical society that speaks on behalf of nature. As a consequence of this general attitude towards critique, Latour will go on to argue, in his “Compositionist Manifesto,” that modernist critique relies upon “the most contradictory notion of an action without agency,” or he would argue that normative standards are inseparable from a multiplicity of heterogenous historical factors and their causal effects. But I suspect that the form of deontological critique Latour criticizes is not exactly how you are developing it, especially if you include Nietzsche and Foucault as inheritors of the Kantian critique. I am intrigued by your project, especially since I’ve been working through Davidson and have some familiarity with Brandom, so it may well turn out in the end that we both largely agree on these matters. That said, when it comes to the normative standards one can find in Latour, I think one would need to look towards an Aristotelian virtue ethics rather than a Kantian deontological ethics. I have tried to sketch out how this might look in a previous post – my Adam Smith post – but I anticipate your follow up claim, which again I can see in Davidson as well; namely, in order to even begin determining the right decision from a virtue ethics perspective, or the right composition from Latour’s perspective, one must already have a basis upon which to determine what it is that is being decided and how one would then determine the success or failure of the decision. As long as this deontological basis is itself open to transformation – or, as the analytic tradition might put it: as long as it doesn’t resurrect the analytic-synthetic distinction – then there may well be a way to reconcile a virtue ethics with the deontology that is for you a first science. I’ll look forward to seeing how your arguments develop.

  3. Pingback: Deleuze – the new unfalsifiable academic guru? « Archaeological Haecceities

  4. It’s possible that I find the OOO and Networkologies projects so intoxicating because of all the effort and hair pulling I’ve done in the last 10 years as decades of foreign policy are unraveling and speaking plainly seems to have no effect on those who need to hear the truth most. You know, neighbors.

    I don’t know if any of you realize just how important the work you are doing really is.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s