Reducing the principle of Irreduction

As Graham Harman (2009) pointed out, the second part of “The Pasteurization of France (Latour, 1988), Irreductions, is perhaps the most philosophical work of Bruno Latour: ” ‘Any argument about my “philosophy,”’ Latour writes, ‘has to start with Irreductions, which is a totally orphan book.’” (Harman, 2009: 12). The philosophical content of Irreductions is also another way of approaching a sort of object-oriented empirical philosophy: “Latour always insists that we cannot philosophize from raw first principles but must follow objects in action and describe what we see. empirical studies are more important for him than for almost any other philosopher”. (Ibid: 14).

Latour’s object-oriented philosophical proposal revolves around what he calls the principle of irreducibility. This principle claims that: “[n]othing is, by itself, either reducible or irreducible to anything else […] but it is a prince that does not govern since that would be a self-contradiction.” [and he continues] “There are only trials of strength, of weakness. Or more simply, there are only trials. This is my point of departure: a verb, “to try.” […] It is because nothing is, by itself, reducible or irreducible to anything else that there are only trials (of strength, of weakness). What is neither reducible nor irreducible has to be tested, counted, and measured. There is no other way. […] Everything may be made to be the measure of everything else.”

In 2013 Levi Bryant started a discussion on his blog (Larval Subjects) criticizing himself for having written an article defending Latour’s principle of Irreduction two years earlier. “Having reflected on this principle in the intervening years, I can’t help but believe that it would be a catastrophe to any knowledge-producing practices were it taken seriously.  Why?  Because [according to him] to explain is to reduce.  The sciences explain the powers of H2O by reference to the features of hydrogen and oxygen.  Likewise, I explain the powers of hydrogen and oxygen by reference to more elementary particles.”

Bryant continued: “The problem with the principle of reduction, when taken at face value, is that it leads us to treat every entity as an ontological given.  ‘God, is but a set of beliefs, you say?  Well by Latour’s principle of [ir]reduction this is an illegitimate reduction!  Therefore we must include God in our ontology!’  ‘Your depression is a chemical imbalance, you say?  Well that’s an illegitimate reduction and it really means all that your confused says it means!’  And while you’re at it, us Jews really were what the Nazis said we were because, well, it would be reductive to say otherwise!”

The answers to Bryant’s thoughts on reduction were going in the opposite direction. For example, a user, Craig McFarlane, remarked: “I don’t buy this line of argument. If an explanation is a reduction to a lower level, then the only possible explanations are those done in terms of elementary particles. But this is ludicrous. The point about H20 (or organizations or corporations or religions) is that they have emergent properties that they would not have were they not organized in such a away. To combine those parts (two hydrogen and one oxygen) in such a way produces a certain thing that is not reducible to two pieces of hydrogen and one piece of oxygen. After all, if you put just oxygen and hydrogen in the proper ration of 1:2 in a bottle, you don’t get water. Likewise, if you put a bunch of accountants in a room you don’t get an accounting firm.”

Michael Norton (another user), implicitly proposed there are two kinds of reductions, good ones and bad ones: “If water is an emergent entity with unique powers, why not a rainbow? Why not a god? Provided they are fabricated well, according to their respective felicity conditions. The difference between a good reduction and a bad reduction can’t be established simply according to type, especially if what matters most is that we show our work in each case.” In the same line, a bit exalted, user Philip complained about Bryant’s notion on reduction and how he misunderstood “a quite simple principle”:

“Latour is never saying that we must stop ‘reducing’ things altogether. That’s the criticism that’s usually leveled at this principle and it’s nonsensical. Ray Brassier’s ‘critique’ of Irreductions in the Speculative Turn is based almost entirely upon this elementary misunderstanding. […] Reduction is more or less a *synonym* for relation – but you can only relate two things that are irreducible to each other (since otherwise there would not be two things at all). […] The argument that science can only work by reduction fits with Latour’s axiom perfectly well. To reduce is to form a network. There’s no issue. But while science may ‘reduce to explain’ it doesn’t ‘reduce away’ as in ‘explain away’ – there’s always a remainder.”

Thus, good reductions are those descriptive —ethnographic— exercises in expanding and relating elements (epistemological reductions) instead of just explaining them (theoretical reductionism). The trick has always been the same: following the actors and letting them talk when one is mapping their trajectories, encounters and disagreements, in a situated way but from many different places and using many different lenses. The point here is to conceive of public spaces (regarding our interests in the linkage between Urban Studies and STS) as actants, as temporal multiplicities composed by any kind of relations, avoiding the classical glance of considering them as solid and stable locations —stages— where social situations happen.

In other words, public space is never a singularity or (only) a particular element or scenario. Instead, Times Square, for instance, is a sort of collection, a kind of palimpsestic sum of situations and versions of that ideal and temporal stable element we know as Times Square. That momentary stabilization is more a product of our efforts in approaching reality than a particularity reality has. So, to summarize, when we talk about reducing an actant, we are tracing and disassembling a multiplicity — a process for which Deleuze used the rhizome as his epistemological tool whereas Borges used the concept of the labyrinth— not with the idea of explaining it through the application of ready-made theories but by showing how that multiplicity is constantly composed and recomposed.


Harman, G. (2009). Prince of Networks: Bruno Latour and Metaphysics. Melbourne:

Latour, B. (1988). The Pasteurization of France. Cambridge: Harvard University Press.