Situated Symmetry: resolving epistemological positions between ANT, OOO & Phenomenology
doctoral marginalia
In recent years, both flat ontological and phenomenological approaches have become more prevalent in ethnographic research. However, it is not always understood in the research communities that these positions are antithetical to each other. Sometimes one will even find both approaches used in the same article (e.g. Roe 20016). Don Ihde (2002) addressed this incompatibility in his essay, “You Can’t Have it Both Ways: Situated or Symmetrical.”
Here I will argue both with and against Ihde, and claim that while these positions are not resolvable on the terms of either discourse alone, they can be integrated in a larger perspective culled from Maturana and Varela’s biological phenomenology, where they describe the distinctions of 1) knowing from within, 2) describing from within, and 3) describing from without.
Applying this model to Ihde’s contrast between situated and symmetrical approaches, one can assign phenomenology to position 1 (knowing from within) and flat ontology position 3 (describing from without). Moreover, Gerard Genette’s literary theoretic concept of ‘focalization’ — originally developed to describe shifts in narratorial perspective — can be used to describe shifts in perspective so that this incompatibility between flat ontology and phenomenology is removed by a more agile observational method.
Critique of Flat Ontologies: ANT and OOO
Both ANT and OOO are ‘flat ontologies’ in the sense described by Manuel DeLanda (2013):
…while an ontology based on relations between general types and particular instances is hierarchical, each level representing a different ontological category (organism, species, genera), an approach in terms of interacting parts and emergent wholes leads to a flat ontology, one made exclusively of unique, singular individuals, differing in spatio-temporal scale but not in ontological status. (p. 47)
ANT and OOO each ‘flatten’ out of different but related motivations. OOO flattens because everything is a thing, sealed within the vacuum of its individual ‘tool-being’ (Harman, 2002), and does not communicate the full richness of its being to neighboring or adjacent things, whether those things are tool-users or just other things it is in contact or proximity with. OOO originated out of an analysis of Heidegger’s reflection on the broken tool.
Its chief exponent, Graham Harman, extended Heidegger’s broken tool reflection to the entire realm of objects– to briefly summarize, a broken tool reveals to the user that there is much more to the tool than just its practical affordances, and this general surplus of being that is never fully communicated to other beings is taken as a general state of affairs, whether or not a thing happens to be a tool or not, or whether or not there is a user around.
ANT flattens because of its conceptual rootedness in the semiotic equivalence of all signs which belongs to the Saussurean tradition of semiotics, which underpins post-structuralist theoretical orientations in general. In this order of equivalence, all signifieds are framed as actually consisting of other signifiers and thus there is no ‘real’ signified but only an continuous loop or chain of signifiers pointing at other signifiers in an infinite regress.
The semiotic perspective employed in my research on multimodal display is Peircean, not Saussurean. In Peirce’s triadic model of Symbol, Index and Icon, the the Saussurean concept of the sign is restricted to the Symbolic dimension, where the symbol takes on meaning only through the rule of interpretation. As we will see, the indexical and iconic relations between signs, referents and interpretants is indispensable for a fuller account of prototyping new media systems. Whereas in postmodern semiotics the discourses often wrote of ‘circulating signs,’ in Latour this model of signification is extended to cover the whole of the material world so that he writes instead of ‘circulating reference’ (1999a).
With its post-Saussurean and poststructuralist epistemic character, ANT considers signifieds to be “essences” shorn of social construction, and thus are to be critically opposed. Even the physical properties of objects are considered to be signified essences and thus banned from the discourse as inherence in things, since all signifieds are considered to be reducible to other signifiers in chains of signifiers-only. Thus in ANT all signifiers are of an “equal” (symmetric, nonhierarchical) status, called by ANT “generalized symmetry,” (Callon, 1986) and thus anything under consideration is framed only as having a capacity to “act.”
At the same time, there are many interesting features of ANT that parallel systems theory. ANT sometimes resembles systems theory without the positivist baggage of the empirical, since “actor-network” is analogous to “system” where “actor” or “actant” corresponds to “part” and “network” to “whole,” so “actor-network” is the dynamic whole that is more than the sum of the parts and so on. Latour on occasion employs one of the favourite tropes of systems theory, the idea of wholes ‘nested’ into other wholes, which in systems theory is a subsumption hierarchy.
In Latour it is not uncommon to find three discursive topologies– chain, nested hierarchy, network– interacting with each other, and perhaps also somewhat contradicting each other at least in terms of their formal topology. Latour’s network is not the network of graph theory or telephone or ethernet networks, or neural nets or network society. Rather, the network is a general trope for something like a ‘web of associations’ or ‘relations between actants,’ which is to say the network of ANT is somewhat only metaphorically a network.
[T]hey can point with their fingers to phenomena taken in by the eye and susceptible to the know-how of their age-old disciplines: trigonometry, cartography, geography. In accounting for this knowledge thus acquired, we should not forget to mention the rocket ship Ariane, orbiting satellites, data banks, draftspeople, engravers, printers, and all those whose work here manifests itself as paper. (Latour, 1999a, p.29).
So the network of ANT indeed can include computational networks, but also the causal relations across time, networks of professions, networks of disciplines that are ‘age-old,’ failed grant applications, and “the indefinite sedimentation of other disciplines, instruments, languages and practices” (p. 30). Circulating reference, in the construction of scientific knowledge, traces these connections whereby “Space becomes a table chart, the table chart becomes a cabinet, the cabinet becomes a concept, and the concept becomes an institution” (p. 36). “Plants are not exactly signs, yet they have become as mobile and recombinable as the lead monotype characters of a printing press” (p. 38).
Yet despite this ever-present expansive image of the network of relations and associations– or we can say, a network expressed in qualitative natural language as sociological theory and not mathematically constrained by the edges and nodes of graph theory, or the signal flow of data cables– the classic systems notion of the nested hierarchy, which was Bertalanffy’s original conception for general systems theory (to array all phenomena as nested systems), sometimes creeps into Latour’s language, even though the index section of his books tend to have no place for systems theory. In addition to this idea of nested structure, one also comes across notions of stages and levels, which is counter-network in its topological effects and is in fact hierarchical language.
Notice that, at every stage, each element belongs to matter by its origin and to form by its destination; it is abstracted from a too-concrete domain before it becomes, at the next stage, too concrete again. We never detect the rupture between things and signs, and we never face the imposition of arbitrary and discrete signs on shapeless and continuous matter. We see only an unbroken series of well-nested elements, each of which plays the role of sign for the previous one and of the thing for the succeeding one. (p. 56, my emphasis)
In Reassembling the Social (2005), Part I of which explicates the ‘five sources of uncertainty,’ Latour writes that “all the sources of uncertainty are nested into one another, that a report written by some humble colleague who does not even where a lab coat may make a difference” (p. 139). Latour’s diagrams often look classically structuralist in their aesthetic, with diagrammatic stylings redolent of hierarchical thinking:
Different from this topology is the chain-structure which is also found in Latour’s diagrammatic and conceptual explorations.
Note, however, how neither of these Latourian diagrammatic examples look anything at all like what is typically understood to be a network:
It is thus somewhat notable when, after rigorous empirical observation of scientists at work, which have yielded diagrams and language that resembles systems theories concepts of nested levels and linear chains– tropes which also sometimes creeps into his own writing– that Latour re-asserts the network language with which he is associated, as though his own thinking hadn’t taken him well beyond the network image:
Constructing a phenomenon in successive layers renders it more and more real within a network traced by the displacements (in both senses) of researchers, samples, graphics, specimens, maps, reports, and funding requests (p. 76).
At times some of the ‘nodes’ of the actor-network take on the appearance of what in intertextuality theory was known as ‘irrelevant intertexts’ since there appears to be no way to distinguish the general fact that everything is related to everything else, in some way or other.
With its metaphoric conception of network, ANT brings all empirical phenomena into a radical semiotics that derives from Saussure’s relational system of negative differences. There is only a set of relations to describe and the method of ANT is the tracing of these relations or associations.
Each level of technology transfer process involves many different entities whom interact to each other to attain their objectives respectively. Entities include individual, groups of individual and and non-human elements that represent various roles, such as technology recipients, technology messenger, employee, management, trainee, trainer, customer, partner and consultant. Non-human elements include technology, machine, business report, document, business strategy, policy, plans and culture. (Kasimin and Ibrahim, 2001, p.195)
Note here how an inanimate object, framed in an ANT discourse, can be said to have “objectives” since any entity in an actor-network is understood to “attain their objectives respectively.” It is also odd that even “culture” can be just another object in the network, like a machine, a policy or a consultant.
As noted in my outline of material practice above, objects are ‘on the wrong side’ of agency. In the frame of ANT, “agency” is simply an empty binary term arrayed against the empty term “structure,” both of which are classic conceptual poles in sociology. In this context my working definition of agency will be simply, “The capacity to make decisions that manifest themselves in meaningful outcomes.” Agency is not a particularly important concept for my research, since it is well encapsulated in the more encompassing notion of virtuality.
The capacity to make decisions presumably has arisen as a form of cognitive emergence by the need of organisms in their environment to exploit the action possibilities of the objects and spaces around them in order to thrive. Thus, an animal may need to determine whether water is over this hill or that, or whether a noise in the trees is a predator or the wind, or whether if something is eaten it will be edible and nutritious, poisonous or inert. Agency is on the side of the decision maker, and the role of objects in the environment is to emit ambient information that is taken up by the neurological reconstruction of a world by a living being which needs to interact with its environment.
But objects, in emitting ambient information (this is something of a phenomenological way of putting it, since there are no transmitters on objects that radiate “info” with invisible antennae), are not agential. Roving life forms construct an order of virtuality out of their local environment, and in so doing produce a ‘lopsidedness’ of agency as a production of embodied cognitive interaction with a world.
This interaction I hold to be asymmetric, since objects could care less and decide nothing. Put another way, objects produce entropy, agents negentropy, and agential negentropy itself is founded upon the negentropy of complex organic systems. Objects do not have “equal status” with people in regards to agenc. They very much do have equal status with people with regards to other things, such as gravity (people and pears and pebbles will all fall at the same Newtonian rates of speed if dropped from a height).
While ANT uses terms “actors” and “actants” rather than “agents,” action possibilities, as discussed by Gibson, are what is taken up by organisms who make use of their surroundings. Objects have “affordances” and action possibilities are asymmetrically on the side of the organism, not its surrounding objects. Actions take up affordances for their realization, and so the actants are not equal, because action is mobilized by the self-organized complex being who interacts with things through their reconstruction in virtuality, which produces information for embodied interactions. Virtuality is the actual as information, and information shapes interactions.
Agency is a property of virtuality, and virtuality itself is the construction or reconstruction of the actual, since virtuality is produced out of the interactive sensorium, in which a sense of the exogenous is created out of endogenous processes. For sure there is much “activity” in the universe, e.g. actions and reactions, solar flares or repetitive orbits, but what is decisive for material practice is not that a tool user is analogous to a chemical reaction, or a tool on the wall is like an inert substance, but rather the meaning and experience of the activity taken up within the virtualized world of the agent.
There is another interesting problematic with the very concept of a network of general symmetry. The solar system isn’t a network nor are tectonic plates. Root systems and nervous systems, and communication media produced by folks with nervous systems, are networks and exhibit essential network features. Networks appear to be asymmetrically aligned in the universe on the side of living organisms, and as such are the basis of virtuality. Note, for example, that the nested hierarchical diagrams often produced by systems theorists make for rather linear representations:
Once life gets into the picture, interacting with its environment, the complexity takes on a more network-like diagrammatic character of many edges and nodes, as in a neural network. Networks are a type of system, but not all systems are networks. Network theories usually involve either the human, biological, or technological domains, and other uses of the term are a bit of a metaphorical stretch, since neither the solar system nor the galaxy nor atoms nor sparky quarks are anything at all like networks.
Because the phenomena of networks are empirically asymmetric with respect to the rest of the universe — existing only at a certain level of organized complexity: life, nervous systems, culture, communication technologies– one could argue that ANT in fact cannot achieve its radical general symmetry by reducing all phenomena to actor-network semiotic systems. ANT may be “faithful to the insights of ethnomethodology” (Latour, 1999b, p.19) but it also exhibits poor phenomenology– which is also just to point up the difference between an outside observer’s account versus a first person account.
ANT’s key concept of ‘general symmetry’ is ironically asymmetrical because network phenomena are one-sidedly associated with human entities –their brains, bodies, language and symbolic systems– and their technically extended activities. That networks don’t exists beyond the biological calls into question the flattening of nonhuman entities into the same topological scheme of network-made and network-producing makers.
Tools I use are not part of my professional network, though do belong to my network of meaning, which is reducible to the fact that I have a nervous system and my fellow beings that have similar ones have formed societies and cultures. But the materials we use to construct our built environments are not networked in character, even if we use rhizomatic wood or make inventories of them on the internet, because the trees are only part of the tree network, or information about things are part of cyberspace. As co-creators of our atmospheres, plants have a systemic relation to out bodies of course, but that systemic relation is not networked, but rather composed of many dynamic and chaotic forces, not simply nodes, edge counts and their relations which are what networks essentially are.
This of course is not the place to initiate an in depth engagement with ANT or graph theory (which is the mathematical formalization of networks), however given its prominence in some disciplines it is worth distinguishing my view of material practice from this widely held view of networks that horizontalize (equalize) human and non-human actors. There is a fundamental and empirical asymmetry in the relation between makers and materials, an asymmetry which ultimately has its origins in complex organization, formations of entropy and negentropy, self-regulating adaptive systems, and the other phenomena explicated by a variety of systems theories.
Decisive for material practices are virtuality and actuality, motivation and causality. Configurations of networks, actants and actors may very well be a potent way of analyzing the sociological dimensions of complex technological projects, but these projects build upon the underlying character of material practices– as actuality / virtuality– which are not confined to or fully explicated by actor-network analysis.
On the other hand, some concepts from ANT are appealing. For instance, I would be happy to call myself a “heterogeneous engineer” and a “System Builder.” Also, what I will call “transdiscursive” in ANT is known as “translation” which “appears as the process of making a passage between two domains, or simply as establishing communication.” (Brown, 2002, p. 3) “Translation involves creating convergences and homologies by relating things that were previously different.”(Callon, 1981,p,211) However, an account from the inside of the making process will yield phenomenological accounts that contradict ANT, such as the asymmetry of agency discussed above.
There are two final reasons to note ANT’s inapplicability to the material practice of prototyping. First, it shares a common feature with much humanistic critique in that it is intentionally designed to occur long after the fact of its object of study being established:
[I]nstead of taking a reasonable position and imposing some order beforehand, ANT claims to be able to find order much better after having let the actors deploy the full range of controversies in which they are immersed. It is as if we are saying to the actors: ‘We won’t try to discipline you, to make you fit into our categories; we will let you deploy your own worlds, and only later will we ask you to explain how you came about settling them.’ The task of defining and ordering the social should be left to the actors themselves, not taken up by the analyst. This is why, to regain some sense of order, the best solution is to trace the connections between the controversies themselves rather than try to settle any given controversy. The search for order, rigor, and pattern is by no means abandoned. It is simply relocated one step further into abstraction so that actors are allowed to unfold their own differing cosmos, no matter how counter-intuitive they appear (2005, p. 23)
Prototyping is immersed in its present, “unfolding” its “own differing cosmos.” Much of the critique of ANT here is also just to say, of course, that prototyping is not sociology! It literally makes no sense to perform ANT while prototyping, since ANT intentionally waits until all the dust has settled before creating its map of associations. ANT is done after the fact of some new production.
The final point of contention is with respect to the multiplicity of things themselves espoused by ANT, which emerges from its analysis of controversies and “natural objective matters of fact” (Latour, 2005, p.116):
This has nothing to do with the ‘interpretive flexibility’ allowed by ‘multiple points of views’ taken on the ‘same’ thing. It is the thing itself that has been allowed to be deployed as multiple and thus allowed to be grasped through different viewpoints, before being possibly unified in some later stage depending on the abilities of the collective to unify them. There are simply more agencies in the pluriverse, to use William James’s expression, than philosophers and scientists thought possible. (Ibid.)
My contention earlier is that the actual is strict and literal. My evidence for this claim is technology itself, which is the material implementation of virtual mathematical theories transposed into the material orders of the actual. The only way that my cellphone could pick up a signal, or a plane be kept aloft, a genome edited, an atom bomb detonated, a headache soothed or a sine wave produced by software, is if the actual were strict and literal. If it were not strict and literal, then mathematics– the very language of strict literality– would have no purchase on it, and all technologies would fail continuously.
In virtuality is the freedom to be non-strict non-literals, which is just my formally odd way of saying being human. The world of made things in which we live superimposes and negotiates our virtual freedoms into the causal chains of strictly literal materials. Thus, the experience of working with technologies, when its mathematical ontology is thought through, denies this idea that things in themselves are plural and multiple, and in fact subsumes controversies into the sphere of interpretation or even into the sphere of making, since laboratories are full of made things which are used to create discourse.
The human–world correlate is only one of trillions of thing–world correlates. (Morton, 2013, p.227).
Paintings have always been made of more things than humans. They have been made of paint, which is powdered crystals in some medium such as egg white or oil. Now when you put the painting on the wall, it also relates to the wall. A fly lands on it. Dust settles on it. Slowly the pigment changes despite your artistic intentions. We could think of all these nonhuman interventions as themselves a kind of art or design. Then we realize that nonhumans are also doing art all the time, it’s just that we call it causality. But when calcium crystals coat a Paleolithic cave painting, they are also designing, also painting. Quite simply then, the aesthetic dimension is the causal dimension, which in turn means that it is also the vast nonlocal mesh that floats “in front of” objects (ontologically, not physically “in front of”). (Ibid., p.24)
We saw that in the ANT framework, any entity– human or nonhuman– was said to have objectives, and in OOO we find similar meaning-effects, e.g. the idea that substances create art. ANT and OOO are interesting discursive correlates, complementing each other along a clear subject/object faultline, with ANT skewing in its interests and foci toward the subjectivity of social formations, and OOO skewing toward objectivity with its placing of “things at the center” (McLuhan, The Dew Lab, 2012) of inquiry.
Despite the ontological flattening each discourse proposes, in fact the Subject/Object binary reappears or is re-inscribed through these two identifiable discursive skews of ANT and OOO in their discursive tropes and concerns, where ANT expands towards complex social webs, and OOO contracts around the singular vacuum sealed uniqueness of noncommunicating tool-beings.
There are other ways in which OOO’s framing of objects as ‘withdrawn’ is also problematic. In material practice, all objects are available for sensorimotor interaction and from there, reflection and imaginative meaning-making. In making, one seeks out the affordances and properties of objects, and configurations of these across multiple objects in order to create systems of objects, and neither the Heideggerian hidden or Harmanesque withdrawn seems very relevant.
This isn’t to suggest that if one is strictly sticking to philosophy, one should not contemplate all that is hidden, withdrawn, and ‘in-itself’ in objects, however in making one just grabs them and starts working with them, using or playing with them, to see what they can do, and how they interact in the midst of a project coming together.
There is another systems-based critique that can be leveled at flat ontology. Bertalanffy originally developed systems theory in order to overcome the intellectual division of his time between vitalism and determinism, and sought in the articulation of general laws of organization a way beyond either reductive causal laws or a mysterious elan vitale of objects.
The object-centredness of ANT and OOO — that the world is just full of things, and we are composed of relations between things amongst things — can be understood as a rejection of the concept of organization as its own principle. That objects are arrayed into systems, subsystems, complex, chaotic or adaptive systems is all ignored by both ANT and OOO, since in flat ontology there are just things, and things tend to be nouns and not organization in general. The concept of thing is extended so widely as to even include a ‘thing’ like ‘culture’ so as to lose any conceptual definition, and in OOO even a relation is thought of as an object.
It is true that systems theories often produce hierarchical models of organization, which humanists often do not like because it looks like bureaucracy. However, these hierarchies found in phenomena are part of the empirical methodology of modelling these phenomena in the most correspondent way that they can be modelled.
Object-centered ontology is not very compelling theorizing for an account of prototyping, especially when what is being prototyped is systemic through and through, intertwining computational, electrical, nervous, haptic, auditory, visual and other systems which exhibit hierarchical and organizational features that are, in systems theoretic terms, function-structures or form-processes, i.e. neither part nor whole– nor just ‘thing’ — but organized.
One might claim that organization is just another thing — because it is a noun perhaps — but systems are wholes made up of self-interacting parts which exhibit emergent features at another level other than the constituent part, which violates the principle of noncommunicating things sealed up in their own ontological tool being vacuums.
Below I will follow Harman’s founding moves in the elaboration of flat tool ontology, as initially developed in Tool Being (2002)
Let’s imagine that, for whatever reason, some sort of bulky metallic appliance is abandoned on a frozen lake. For now, I see no reason to accept the animistic claim that such a stove or washing machine “perceives” the lake in the usual sense. Even so, some sort of determinate encounter clearly does occur between them. This soulless piece of metal certainly does not enjoy immediate and intimate contact with the tool-being of the lake, as if sheer causal proximity were sufficient for capturing that lake in its withdrawn execution….Even in this case, the appliance reacts to some features of the lake rather than others — cutting its rich actuality down to size, reducing it to that relatively minimal scope of lake-reality that is of significance to it.
Note that the tool-being of the lake comprises an indefinitely large array of features, most of them irrelevant to the object lying on its surface. Simplifying somewhat, we can say that the stove reduces the lake to the single aspect of a frozen surface, to sheer “equipment for remaining stationary.” This sort of analysis is familiar enough to readers of Heidegger, at least in a human context. When the lake supports the appliance, this act of supporting unfolds entirely within the as-structure, not within the kingdom of tool-being. This raises the following question: if the fact that the frozen lake supports an object is not its tool-being, then what is? To explore this question, imagine that the ice now cracks or melts, so that the appliance smashes through the surface and sinks into the frigid depths of the lake.
I hold that the resulting interaction between stove and ice is philosophically identical with the more familiar case of Dasein and the broken hammer. For what is decisive in the famous account of the “broken tool” is not that implicit reality comes into conscious view, as if human surprise were the key to the reversal within being. Rather, the important factor is that the heavy object, while resting on the ice as a reliable support, did not exhaust the reality of that ice. The appliance could have been resting either on thin ice or on an eternal pillar of granite, and the supportive effect (prior to the disaster) would have been precisely the same, ignoring for now the specifically icy experiences that the appliance may also have undergone.
The same would hold true for any object in such a predicament: none of the entities near the lake are in position to sound out every last fugitive echo of its being. Just like explicit perception, causal reaction is always only a response to a limited range of factors in the causative entity; other features are passed over, concealed from the object that runs up against it. When one billiard ball strikes another, it treats its victim as a simple mobile mass, and remains unattuned to its other concealed treasures — the richness of its imperfect plastic texture, its suddenly irrelevant color or its vague synthetic fragrance. No object ever unlocks the entirety of a second object, ever translates it completely literally into its own native tongue. (p.222–224)
Flat ontological descriptions invite anthropomorphisms at every turn. As readers we can oscillate between a generous forgiveness– citing the limitations of our language for expressing extreme thoughts– or we revert to a critical attitude and begin to find fault in the extreme conceptualization itself. Here Harman is essentially giving an account of affordances between inanimate objects, since the appliance on a frozen sheet of lake ice is realizing an action possibility in the interaction between objects in an analogous way that any creature can walk on anything.
The difficulty for the concept, however, is that neither object is a perceiving being reconstituting its environment in a field of ambient information which is used to inform decisions about its possible actions. Moreover, only such beings perceive qualities, or what Harman calls “concealed treasures” which are essentially qualia: “the richness of its imperfect plastic texture, its suddenly irrelevant color or its vague synthetic fragrance.” And these qualia– which do not exist for inanimate things — in turn do not go into the objects’ “own native tongue” because of course they do not have language.
What distinguishes the ontological topology of OOO from ANT is the former’s insistence on the vacuum-sealed character of objects, which is as far one can go in the opposite direction from the image of a vast and dense network of relations. The object of ANT is vacuum-packed and remarkably, even relations are said to be just other objects themselves situated in a vacuum.
Just as the steam engine is made up of dozens of elements, silver simpliciter is not just a black box, but also the formal union of trillions of minuscule internal organs. At the same time, it would be wrong to conclude that every entity is only a relational effect, only a network that is fully deployed in its current state. This is not the case. Despite their status as machine-like integrations of countless distinct components, the silver and the engine both have an inherent unitary reality, one not exhausted by the fact that the silver is “currently floating” or that the engine is “currently powering a Chinese warship.” ….
[N]o entity ever exhausts the reality of another, never makes contact with the darkest residues of its heart. Far beneath any prehension of the silver or by the silver, there is the silver in its real execution, silently resting in its vacuum-sealed actuality (a.k.a. “vacuous actuality”), waiting to inflict injuries or blessings on entities yet to arrive. (p.283)
there is no such thing as a sheer “relation”; every relation turns out to be an entity in its own right. As a result, there is no cleared transcendent space that gains a distance from entities to reveal them “as” what they are. There is no exit from the density of being, no way to stand outside the brutal play of forces and vacuum-packed entities that crowd the world. We ourselves are only one such entity among innumerable others. (pp.289–290)
Although Whitehead employs the term “vacuous actuality” as an insult, it can be salvaged as a neutral description for the true reality of objects. The actuality of the object belongs always and only to a vacuum. (p. 290)
The tools withdraw into a vacuum, an “extraworldly” refuge. But where might this refuge be? On the one hand, tool-being commits us to the existence of tools in a vacuum. On the other hand I have argued that there are no gaps in the cosmos, that the world is stuffed absolutely full with entities. But this seems to suggest that although the tool-being of a hammer withdraws into a vacuum apart from any relations, this vacuum can only be the body of another entity, since there are no other kinds of places. (p.294–295)
However, eventually in this discourse the sheer difficulty in imagining these pure vacuums not only of all things but of all relations (which are also taken to be things) seems to require the re-entry of some kind of causal relationship to be brought back into the picture, which Harman calls “occasional cause.”
Given that direct ontological contact between substances now seems impossible, the time has come to revive some form of occasional cause. (pp.295–296)
The occasional causality I have in mind would have to occur on a more local level, and not away on high in the sphere of the divine. If a faulty electrical cord sets fire to a rug, even though the withdrawn tool-being of these objects never really comes into contact, then we must ask through what medium they affect one another at all. And in a strange sense yet to be determined, that medium will have to be another object , because according to the results of this book there is nothing in the universe but objects. (p.296)
In any case, we are left with the following scenario — the world as a duel of tightly interlaced objects that both aggrandize and undermine one another. The movement of philosophy is less an unveiling (which relies on an illegitimate use of the as-structure) than a kind of reverse engineering . Teams of industrial pirates often lock themselves in motel rooms, working backward from a competitor’s finished product in an effort to unlock and replicate the code that generates it.
In the case of the philosopher, the finished product that must be reverse-engineered is the world as we know it; the motel room is perhaps replaced by a lecture hall or a desert. Behind every apparently simple object is an infinite legion of further objects that “crush, depress, break, and enthrall one another.” It is these violent underground currents that one should attempt to reverse-engineer, so as to unlock the infrastructure of objects. Whatever the details of its functioning may turn out to be, this infrastructure must be made up of tools in a vacuum (p.297)
It will be clear that such a conception makes no sense for actually making anything. Objects are brought into causal relationships and intentional configurations without any concern for what may be going on at their supposed “heart” or with what “enthralls” them. There are also clear practical distinctions to be made between what is an object, a signal, a relationship, a cause or effect, an action, an idea and so on that calling these all by the same moniker– “object” or “tool being”– clarifies nothing for the making or the maker.
There is a way that ANT and OOO make for an interesting complementary conceptual scheme, as this skew that has been identified– either toward the web of relations, or toward the inwardness of things– can be combined to offer a kind of metaphysical version of graph theory, which is the formalization of networks.
In this complementary and hybrid view, OOO provides the metaphysical theory of the node– the withdrawn object full of untapped and uncommunicated qualities and being– while ANT provides the metaphysical theory of pure relations– the dense web of everything connected to everything else. Taken together, one gets two theoretical lenses that can be superimposed on each other to provide a node-and-edge account of possibly everything. Let’s take this one step further. Networks are sometimes opposed in theory against hierarchies.
Networks are not planned by a single authority; they are the main source of innovation but are relatively fragile. Hierarchies exist primarily because of economies of scale and scope, beginning with the imperative of self-defense. To that end, but for other reasons too, hierarchies seek to exploit the positive externalities of networks. States need networks, for no political hierarchy, no matter how powerful, can plan all the clever things that networks spontaneously generate. But if the hierarchy comes to control the networks so much as to compromise their benign self-organizing capacities, then innovation is bound to wane.(Ferguson, 2014, n.p.)
Unsurprisingly, networks are often portrayed more positively compared to negative (always bureaucratic) hierarchies. Yet these are not clearly opposed but often integrated topologies.
If we imagine a collection of companies linked by communication networks, we can find hierarchies not just at the site of the node (e.g. in a university, this may be a hierarchy of associate, assistant and full professors, or in a company lower, middle and upper management), but the communication network, as an ensemble of engineering artifacts, is hierarchical as material technical systems, whether it is the levels of abstraction of software– machine language, firmware, assembler, kernel, operating system and applications — or at the level of hardware e.g. in the physical stem hierarchy– discrete components, printed circuit boards, integrated circuits, functional modules, electronic then electro-mechanical systems on a chip, and so on. In this concrete example a node is a social hierarchy and an edge a material hierarchy.
Indeed, such a discursive conception– of ANT and OOO together providing the philosophical version of graph theory– can also be a kind of systems theory, since in the example above– social organizations connected by communication networks– the network would be a system-level ‘higher’ than its component ‘lower level’ hierarchies– but together this nesting of node and edge hierarchies into a network topology is itself a “subsumption hierarchy” to use a term from systems theory, since edge and node hierarchies– companies and communication systems– are component parts of the larger network system whole.
If ANT tends to give the theory of pure relations (of objects), and OOO tends to give the pure theory of the object (non communicating very well with other objects), this proposed combination of the two theories at least provides some way to move forward toward more recognizably empirical phenomena.
However, it is interesting to note that in the ‘real world example’ offered– of companies linked by communication networks– the social hierarchy is at the level of the node while the physical hierarchy is in the edge. This amounts to a complete inversion of the philosophical mapping of ANT and OOO whereby the node is the physical object and the edge is the social relation.
It is not the task here to explore the question of ‘what to do’ with this hypothetical and proposed mapping and inversion of graph theory across actual and virtual domains, but only to note some interesting topological patterns that emerge when considering ANT and OOO together in their relatedness as flat ontologies.
This critique of flat ontology is entailed by phenomenological method in general, as poststructuralist perspectives are explicitly anti-phenomenological in character (e.g. no signifieds or correspondences allowed). Don Ihde has expressed the postphenomenological position — where the “post” here means moving beyond the traditional focus on embodied experience to include technically mediated perception and experience– vis-a-vis flat ontology as follows:
I am opting for a perspectival, situated knowledge which lacks the god’s eye view either from overhead or into the interior. Yet I also do not want to make the symmetrist’s equivalent error: simply granting some kind of equality of status to the human and the non-human. On the one hand, we have seen this leads either to the temptation to “mechanize” the totality or to “socialize” it.
There is simultaneously in the modernist and the symmetrist’s positions the temptation to a kind of reduction in one or the other direction. Rather, and this has been my point, an asymmetrical but post-phenomenological relativity gets its “ontology” from the interrelationship of human and non-human. (p.143) [original emphasis]
References
Ihde, D. (2002). Bodies in Technology. University of Minnesota Press.
Roe, E.J. (2006). Things Becoming Food and the Embodied, Material Practices of an Organic Food Consumer. Sociologia Ruralis, Vol. 46, №2. pp.104–121.
