, New Urban Conditions

Challenging the Assemblage Dictum of the (Post-)Anthropocene

Author: Valerii Shevchenko

Moving beyond the limits of assemblage theory, thinking of the Anthropocene as a system that can be observed and studied, can help us discover new political horizons—concrete solutions to the challenges of our time.

Debates over the existence of climate change as the most evident process of the Anthropocene have an unavoidably political character. The fact of climate change confronts both nations and humans with choices. In parts of the world, national politics are now at least partly based on ecological agendas. Other states and policy-makers deny the existence of climate change, presenting it to the public as a tool used by rival countries to weaken their economies.

In other words, there is strife around the political implications of climate change and other anthropogenic influences on the Earth. But, across many natural disciplines, the evidence is clear: there are significant shifts in the planet’s biogeochemical systems. It is vital, then, to acknowledge the Anthropocene as a new geological epoch and to comprehend its philosophical implications.

The question of agency is central in the philosophical understanding of the Anthropocene. Many thinkers try to reinvent epistemologies and ontologies to escape from the Cartesian dualism of res cogitans and res extensa and to break down the commonsensical dichotomies of subject and object, nature and culture, and others which are implied by the dualist paradigm. The idea behind such moves is that dichotomies are analytical accomplishments and therefore not quite “real.” As it is said, any dichotomy has some ground on which the very distinction is made and which needs to be accounted for.

There are many approaches to such reinvention: actor-network theory (ANT), new materialism (NM), object-oriented ontology (OOO), multinatural ontologies, and other theories. All aim at decentering the human position. In this view, the human is one object among many, not a central organizing principle of the universe. All these approaches are concerned with ontology. They prioritize questions of being and of agency, asking: What does it mean to be and to act beyond the human-centred perspective based on Cartesian dualism?

This is not to say that these theoretical approaches arose in response to the problems of the Anthropocene, or that they are somehow indebted to it. Almost all these approaches existed before the articulation of the concept. But the arrival of the Anthropocene and the explanation of its vital significance to life on Earth reinforced the agendas of post-dualist philosophical approaches.

The humanities underwent a turn to ontology in the 1980s and 1990s in the works of proponents of ANT and NM. The Anthropocene was scientifically elaborated in the early 2000s. Due to the reinforcement of initial ontological programs in humanities, anthropologists Martin Holbraad and Morten Petersen—who pioneered in the ontological turn in anthropology—say that in today’s philosophical discourse “…new ontologies now abound, in the form of an array of novel conceptual vocabularies and aesthetics of thought: assemblages, affects, networks, multiplicities; posthumanism, multi-species and the Anthropocene, new materialism, speculative realism; emergence, vibrancy, intra-action; the para- , the off- , the pata-.”

The concept of assemblage is the most frequently used concept concerning agency in the discussion of the Anthropocene. It illustrates the heterogeneous, distributive, and processual character of agency. Understanding of the causes of the Anthropocene requires, even, a description of the “agency of the assemblage.” However, the concept of assemblage is misleading in numerous aspects that originate from its initial philosophical and epistemological contexts.

Sofa Skidan, Transverse hyperspace series, fragment, 2018. Courtesy of the artist


Scientific and philosophical views of the (post-)Anthropocene

Setting the stage

The concept of the Anthropocene was proposed by atmospheric chemist Paul Crutzen about two decades ago. It represents the geological epoch defined by anthropogenic causal factors affecting the Earth as a single, evolving planetary system. One of the defining factors is the high level of carbon dioxide (CO2), which affects energy balance at the surface of the Earth.

Beyond climate change, humans are responsible for affecting other biogeochemical cycles such as nitrogen, phosphorus, and sulphur cycles which are fundamental to life on Earth—as well as having an impact on a range of environmental systems. All those processes are evidence of the impact of human activity on nature which humans are now faced with.

The situation of the Anthropocene characterized by vastly increased values of biogeochemical indicators in quite a short period of time almost forces humans to rethink their own position relative to nature. The question of causes of all those changes leads to the idea of rethinking human impact on the environment expressed in their actions and of questioning the very capacity to act as a unique human feature. That is why agency becomes a focal point in the philosophical comprehension of the Anthropocene.

The concept of the post-Anthropocene was proposed by various thinkers, including design theorist Benjamin Bratton, who underlines the idea of independently existing “technologies-for-themselves” forming landscapes of automation and human exclusion zones where humans not only need not be, but where they are not even supposed to be. This concept might be said to be detached from the initial biogeochemical sense pointing to some new geological epoch. It has no natural-scientific elaboration. The post-Anthropocene is not so much an epoch of posthuman agency, but some condition of total indifference of an environment that does not even presuppose humans.

Of course, this is more about architecture and design and not about substantive ontological questions which are discussed in the discourse on the Anthropocene. Nevertheless, post-Anthropocene discourse certainly has its political implications which are to be elaborated in future research.

Turn(s) to Ontology

Digging deeper into our philosophical understanding of the Anthropocene, we need to notice that the main idea behind the decentralization of humans is the commonsensical dichotomy of Nature and Culture (or Society), which is responsible for concentrating all agency and capability in humans—and marshaling it for their benefit. Philosopher Bruno Latour, like others, has suggested that this dichotomy is artificial. It is produced, Latour says, through the double operation of “translation” and “purification.” Translation is responsible for blending different heterogeneous entities into entirely new strange hybrids of nature and culture, while purification is responsible for sorting those into two distinct ontological zones—the human and the non-human. Seen in this light, the object of science is naturally “out there,” and there are many different perspectives on it, be they representations, theories, or something else. Latour calls this ontological stance of one-nature-many-representations “the modern constitution.”

Initially, the idea of artificially sorting entities based on whether they have agency did not relate to the discourse on the Anthropocene. Such arguments were proposed in different disciplines and fields during the 1980s and 1990s and eventually appeared under the label of the “ontological turn.” Strictly speaking, there are a few different turns, each with their own specific characteristics, but we are interested in those in philosophy and science and technology studies (STS), since they relate to the discourse of the Anthropocene.

As anthropologists Holbraad and Petersen point out, the main difference between these ontological turns is that the latter might be said to ask “ontological” questions without taking “ontology” as an answer, while the former does exactly the opposite by establishing metaphysical systems of “what the world is really like.” STS does not take “ontology” as an answer in the sense that it tries to overcome the one-nature-many-representations regime by accounting for the multiple ontologies of the world which are enacted differently in different circumstances. As STS scholars Steve Woolgar and Javier Lezaun note: “It is an effort to circumvent epistemology and its attendant language of representation in favor of an approach that addresses itself more directly to the composition of the world.”

However, “composition” here does not mean any ultimate ontological truth due to STS’s attention and even commitment to the local character of techno-scientific practices. This stance is expressed very aptly by Holbraad and Petersen: “Differently configured localized practices will have different ontological effects, bringing forth different kinds of entities, with no overall ontological scheme to sort them out in a unified way.”

In other words, the goal is not to discover some ultimate and universal ontological truth about the world, but to discover the ways in which the world is composed in local circumstances.

In contrast, the proponents of the so-called philosophical ontological turn view the composition of the world in metaphysical terms. By considering the problem of agency as something that is more-than-human, they try to reorient ontology towards material objects, animals and other non-human actors. Theoretical logic of those philosophical approaches which include actor-network theory, object-oriented ontologies, new materialism, multinatural ontologies, and some other approaches is very close to the one proposed by Latour—to question the distinction of nature and culture and all the binary oppositions implied by it. The difference is in theoretical vocabulary with which the point is made.

The most influential resources drawn upon in elaborations of ontological turn in philosophy are the “machinic” ontology of Gilles Deleuze and Felix Guattari and the relational processual ontology of Alfred North Whitehead. Of course, there are completely distinct resources such as phenomenology in some strands of object-oriented ontology, but it is certainly the reception of Deleuze and Guattari which inhabits the Anthropocene discourse today.

To be more precise, their concept of assemblage is used in the discourse to grasp the non-necessary character of the connection of its parts. Namely, the concept of “global assemblages” proposed by anthropologist Anna Tsing is used to mean socioecological “constellations” including multinational corporations, development initiatives, political treaties, and other forms of governance.

The irony is that Latour’s idea of the “modern constitution” which has now become some kind of intuitive obligatory passage point for researchers was initially founded in the field of STS by empirical studies. In this way, an idea that was strictly empirical and locally enacted has become metaphysical. The “modern constitution,” then, might be said to be situated on the nexus of two ontological turns—that of STS and that of philosophy which is the initial mainstay of ontology.

Sofa Skidan, Memory (not)overcast of nothing, fragment. Courtesy of the artist


The search for some ground underneath artificially constructed dichotomies and for ways of monist thinking is very characteristic for the philosophical ontological turn, and especially for NM. This applies to the philosophical conception of the Anthropocene as well, since it consists mostly of ideas of ontological turns. The Deleuzian principle of the univocity of being might be seen as ground for the concept of assemblage, which is very popular in the discourse of the Anthropocene and which will be discussed further on in this essay. As Deleuzian scholar Nathan Widder aptly points out, this principle expresses an “excess of difference that is ‘common’ to all beings.” This is central to the idea of heterogeneity, which is shared between the ontological turns discussed.

The idea of heterogeneity is in some sense axiomatic for scholars of the post-Deleuzian ontological turn, since the difference is not questionable either in empirical investigations informed by ANT or in assemblage theory and other new materialisms. Univocity is explicitly or implicitly metaphysically postulated which means that it is ascribed to things without any empirical check since it is literally beyond experience. The origin from which dichotomies arise is unitary in its heterogeneity and this is crucial for the idea of agency problematized by ontological turn scholars. To put it simply, agency can be found wherever there is a univocal foundation or ground. Hence, every entity—be it human or non-human—possesses agency.

Discussing agency in the Anthropocene, Latour remarks that “to be a subject is not to act autonomously in front of an objective background, but to share agency with other subjects that have also lost their autonomy.” The idea of distributed agency justifies the notion that the Earth has become a “full-fledged actor” in the age of the Anthropocene by “striking back.” Latour goes on: “Existence and meaning are synonymous. As long as they act, agents have meaning.” Hence, existence and meaning might be said to be the functions of action in a mathematical sense: the more agents act, the more they exist and mean. This relation is quite characteristic for ANT and NM. The inseparability of meaning and existence, of meaning and matter, and of materiality and expressivity is due to the intuition of univocity of being. As Latour notes: “The point of living in the epoch of the Anthropocene is that all agents share the same shape-changing destiny, a destiny that cannot be followed, documented, told, and represented by using any of the older traits associated with subjectivity or objectivity. Far from trying to ‘reconcile’ or ‘combine’ nature and society, the task, the crucial political task, is on the contrary to distribute agency as far and in as differentiated a way as possible.”

This same shape-changing destiny shared by all agents is a very monist train of thought—similar to the idea of unitarity in heterogeneity and hence to that of the univocity of being. An implication of this is that if the difference is common to all agents, there is no obvious ground for any stratification of those agents or entities. If there is no specified hierarchy, the way agents are ontologically organized is “flat.” This idea is crucial for almost every approach of ontological turns in philosophy and in STS as well.

Thus, the idea of heterogeneity leads us to the concept of assemblage frequently used in the discourse of the Anthropocene to denote a whole composed of non-necessary parts which might change over time. In the next section, I will analyze this concept and show that emergence implied by it is inconsistent with “flat” ontology since the former presupposes ontological dependence and hence hierarchy. This will mean that assemblage dictum in the discourse of the Anthropocene should be revisited and either upgraded or abandoned.


Assemblages for the (post-)Anthropocene?

Core problems

As the inventor of assemblage theory Manuel DeLanda points out, the concept of assemblage denotes “wholes characterized by relations of exteriority,” which means that a part of any assemblage might be easily detached from it and put into another assemblage. Those relations are contrary to the logically necessary relations of organismic totality. The concept of assemblage encompasses an idea of contingently obligatory relations between the parts of an assemblage which might become such in the course of coevolution.

It is important to notice that there are two ideas which are crucial to assemblage theory. The first is the idea of emergence, which is that property of the whole that is not reducible to the properties of its parts. DeLanda says that the “property of a whole is said to be emergent if it is produced by causal interactions among its component parts.” This means that he not only defines it in terms of the irreducibility of the properties of the whole to those of its parts, but that he also emphasises causality as the main marker of emergence. Thus, he implies that “if causality, then irreducibility and hence emergence,” which is questionable since there is no necessary connection between causality, irreducibility, and decomposability in the parts of an assemblage.

The second idea is that of population thinking, which presupposes that assemblages exist not in isolation but in populations: atoms, molecules, people, languages, cultures, and so on. This idea helps DeLanda introduce another important part of his theory, namely the “flat” ontology of individuals, where every assemblage is a historically contingent individual and there are no classes, species, or genera, only individual entities. For example, there is a population of cities, each of which is a historically contingent entity, and the “city” is not a transcendent class of which the particular cities are just instantiations. The notion of “flatness” comes to underline that if there is an ontology of individuals, there is no hierarchy presupposed and hence all entities are on equal ontological footing and no one is “more important” than another.

Sofa Skidan, Transverse hyperspace series, 2018. Courtesy of the artist

The logic of assemblage theory is then quite straightforward:

  1. Heterogeneous but ontologically equal entities

  2. stabilize in an emergent whole by interacting with each other

  3. and gain causal powers

  4. and the capacity to causally affect its parts.

The social ontology proposed by DeLanda might serve as an example: he starts by saying that a human individual is an assemblage of sensations and experiences. Then there is a social encounter, or interaction, which is assembled from two or more human individuals and the emergent property of which is the system of practical conventions and “appropriate” ways of behaving in certain situations. After a social encounter comes an interpersonal network composed of different social interactions. The emergent feature of the network is its density and capacity to store the reputations of interactants. Next comes an organization which is capable of distributing the relation of power between different networks. After that, there is a city, nation-state, and so on. As have been noted above there is a problem with the concept of emergence and it needs to be analyzed in more detail.

The concept of emergence is used in systems theory to denote properties of a system which are irreducible to those of its parts. The first problematic aspect of assemblage theory, then, is that the inherently systemic concept of emergence is withdrawn from its initial context, where it has some rigid meaning, and put into another one. An assemblage evolving by causal interactions with other assemblages and forming new emergent wholes standing on equal ontological footing becomes a barely possible picture since the concept of emergence is systemic and a system necessarily has interior and exterior parts. Hence, what assemblage theory does might be called the universalization of exterior relations since all interactions are only exterior and contingent and there is absolutely no place for necessary and interior relations in that theory’s worldview. This flaw originates from the metaphysically postulated idea of the univocity of being, which is not explicated by DeLanda but presupposed by him in the idea of heterogeneity and what we have called ‘unitarity in heterogeneity’ expressed by giving agency to some pre-individual entities.

What DeLanda overlooks is the relation of transcendental necessity between different assemblages. This means that not all assemblages can provide conditions of possibility for each other. For instance, for a social encounter to exist, an individual should exist—individuals are conditions of possibility for interaction and not vice versa since an individual has its own conditions. This undermines the “flatness” of ontology proposed by DeLanda since this means that assemblages are not autonomous in a full sense—they have certain conditions of possibility and are those themselves for other assemblages.

The idea of transcendental necessity concerning the emergent evolution of assemblages leads to the notion of ontological dependence, which means that one thing should necessarily exist in order for another thing to exist. This notion of dependence completely dismantles the idea of “flatness” since it presupposes a hierarchy of entities, some of which metaphysically ground others, which also means that things cannot be parts of each other. In other words, some of them with necessity are more prior than others.

Hence, if parts of assemblages are transcendentally and ontologically prior to assemblages themselves, the parts of the parts are prior to the parts themselves. This leads us to the hierarchical composition of an assemblage which starts to look more like a system with both interior and exterior relations.

Thus, the duty of describing the character of the relations becomes empirical and not metaphysical since we cannot in this situation postulate any ontological principle which will govern our research, namely the principle of the univocity of being.

Sofa Skidan, Do you want to find your CHOMMAA?, 2017. Courtesy of the artist

The Systemic (post-)Anthropocene

What does all this mean for our philosophical understanding of the Anthropocene? If we withdraw the concept of an assemblage and replace it with the concept of a system but keep all the insights about the emergence and contingent relations, we will arrive at the point where those systems might be considered in their dynamics. By putting Deleuze and the idea of the univocity of being aside, we will be able to identify both necessary and contingent relations, which is crucial for identifying the structure of the object studied.

The concept of an assemblage lacks this capability of structural explanation and grasps only the contingent character of the relations of its parts. For instance, the concept of “global assemblage” as a socioecological constellation being replaced by that of a “global system” would mean that that constellation would possess a certain structure which is to be discovered, explained, and apprehended. This would both entail and afford a great political duty of searching for objective and mind-independent characteristics of the systems and of combining and “assembling” them in a structural way.

Thinking about alliances for the (post-)Anthropocene, the concept of an assemblage blurs the political element of our investigation by having a certain ideological bias. It does so by presupposing unitarity in heterogeneity, which is purely speculative and not empirically evident. Replacing it with the concept of a system will allow us to make more accurate approximations of the “really real” world and thus to see the horizon of possible political actions.

Cover image: Sofa Skidan, Addiction to becoming, addiction to non-becoming, 2019. Courtesy of the artist (instagram)

Valerii Shevchenko

A philosopher of social science and a musician. He holds an MA in Sociology from the University of Manchester (UK) and teaches at the Moscow School of Social and Economic Sciences (Russia). Valerii’s work is primarily concerned with general epistemological problems of contemporary social science, naturalistic analytic metaphysics, and the evolutionary explanation of metacommunication in humans. He also practices live-coding in TidalCycles and releases his music on Silhouette Tapes, Oblast, Black Milk, and other record labels under the alias ‘dadaisme.’

If you noticed a typo or mistake, highlight it and send to us by pressing Ctrl+Enter.