Immaterialism

Objects and Social Theory

Graham Harman

• Publisher: Polity
• Year: 2016
• ISBN:
  9781509501007
• Edition: 1
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

What objects exist in the social world and how should we understand them? Is a specific Pizza Hut restaurant as real as the employees, tables, napkins and pizzas of which it is composed, and as real as the Pizza Hut corporation with its headquarters in Wichita, the United States, the planet Earth and the social and economic impact of the restaurant on the lives of its employees and customers?

In this book the founder of object-oriented philosophy develops his approach in order to shed light on the nature and status of objects in social life. While it is often assumed that an interest in objects amounts to a form of materialism, Harman rejects this view and develops instead an "immaterialist" method. By examining the work of leading contemporary thinkers such as Bruno Latour and Levi Bryant, he develops a forceful critique of 'actor-network theory'. In an extended discussion of Leibniz's famous example of the Dutch East India Company, Harman argues that this company qualifies for objecthood neither through 'what it is' or 'what it does', but through its irreducibility to either of these forms. The phases of its life, argues Harman, are not demarcated primarily by dramatic incidents but by moments of symbiosis, a term he draws from the biologist Lynn Margulis.

This book provides a key counterpoint to the now ubiquitous social theories of constant change, holistic networks, performative identities, and the construction of things by human practice. It will appeal to anyone interested in cutting-edge debates in philosophy and social and cultural theory.

Part One
Immaterialism

1.
Objects and Actors

This is a book about objects and their relevance to social theory. Since the books in this series are intended to be concise, I have had to omit a great deal that some readers will regard as central. Influential theorists such as Michel Foucault and Niklas Luhmann appear briefly if at all, while Roy Bhaskar and Manuel DeLanda (both personal favorites) lost entire sections during the final cuts. Instead, the first part of the book will focus on Actor-Network Theory (ANT), which I regard as the most important philosophical method to emerge since phenomenology in 1900, and on New Materialism, the school of contemporary thought most often confused with my own position, Object-Oriented Ontology (OOO).

The track record of ANT in dealing with objects is decidedly mixed. In one sense it already incorporates objects into social theory as much as anyone could ask for. ANT offers a flat ontology in which anything is real insofar as it acts, an extremely broad criterion that grants equal initial weight to supersonic jets, palm trees, asphalt, Batman, square circles, the Tooth Fairy, Napoleon III, al-Farabi, Hillary Clinton, the city of Odessa, Tolkien’s imaginary Rivendell, an atom of copper, a severed limb, a mixed herd of zebras and wildebeest, the non-existent 2016 Chicago Summer Olympics, and the constellation of Scorpio, since all are equally objects: or rather, all are equally actors. OOO could hardly be more inclusive of objects than ANT, and in some respects it is even less so. Yet in another sense ANT loses objects completely, by abolishing any hidden depth in things while reducing them to their actions. After all, you or I or a machine are not just what we happen to be doing at the moment, since we could easily be acting otherwise, or simply lying dormant, without thereby becoming utterly different things. Instead of replacing objects with a description of what they do (as in ANT) or what they are made of (as in traditional materialism), OOO uses the term “object” to refer to any entity that cannot be paraphrased in terms of either its components or its effects.

The search for an object-oriented social theory is motivated by the concerns of object-oriented philosophy (Harman 2010a, 93–104). The first postulate of this philosophy is that all objects are equally objects, though not all are equally real: we must distinguish between the autonomy of real objects and the dependence of sensual objects on whatever entity encounters them (Harman 2011). This differs from neighboring theories that grant equal reality, though not equal strength, to anything that acts or makes a difference in the world, with two good examples being the philosophical positions of Bruno Latour (1988) and much later Levi Bryant (2011). It is not hard to name social theorists who cast as wide an ontological net as Latour: Durkheim’s rival Gabriel Tarde (2012) immediately comes to mind. But whereas object-oriented philosophy treats all sizes of objects equally and considers each as a surplus exceeding its relations, qualities, and actions, Tarde grants privilege to the tiniest “monadic” level of entities, while Latour is reluctant to concede more reality to objects than to their effects. (See Harman 2012a and Harman 2009, respectively.)

A good theory must ultimately draw distinctions between different kinds of beings. However, it must earn these distinctions rather than smuggling them in beforehand, as occurs frequently in the a priori modern split between human beings on one side and everything else on the other (see Latour 1993). This answers the question of why an object-oriented approach is desirable: a good philosophical theory should begin by excluding nothing. And as for those social theories that claim to avoid philosophy altogether, they invariably offer mediocre philosophies shrouded in the alibi of neutral empirical fieldwork.

Concerning the question of whether an object-oriented approach is new, it might seem at first that the theme of objects in social theory is a familiar mainstream topic. Science studies as a discipline, and not just ANT in the strict sense, has seemingly bent over backwards to integrate nonhuman elements into its picture of society. Karin Knorr Cetina (1997) has a good deal to say about objects, though her primary interest is in what she calls “knowledge objects,” and in general her objects are chaperoned by human beings rather than existing outside human contact. Consider also the following promotional blurb for the useful Routledge anthology Objects and Materials:

There is broad acceptance across the Humanities and Social Sciences that our deliberations on the social need to take place through attention to practice, to object-mediated relations, to non-human agency and to the affective dimensions of human sociality. (Harvey et al. 2013)

This passage is typical of recent trends in assigning two, and only two, functions to objects: (a) objects “mediate relations,” with the implication that what they mediate are relations between humans; (b) objects have “agency,” meaning that they are important when they are involved in some sort of action. These are the two ostensibly pro-object insights bequeathed by ANT and related schools. Their praiseworthy aim was to free us from an older tradition in which society was viewed as a self-contained realm where humans did all the acting and objects were passive receptacles for human mental or social categories.

Yet these two key points, however welcome by comparison with what came before, are precisely those points where recent theories have not pushed far enough. To say that objects mediate relations is to make the crucial point that unlike herds of animals, human society is massively stabilized by such nonhuman objects as brick walls, barbed wire, wedding rings, ranks, titles, coins, clothing, tattoos, medallions, and diplomas (Latour 1996). What this still misses is that the vast majority of relations in the universe do not involve human beings, those obscure inhabitants of an average-sized planet near a middling sun, one of 100 billion stars near the fringe of an undistinguished galaxy among at least 100 billion others. If we forget that objects interact among themselves even when humans are not present, we have arrogated 50 percent of the cosmos for human settlement, no matter how loudly we boast about overcoming the subject–object divide. A truly pro-object theory needs to be aware of relations between objects that have no direct involvement with people. This brings us, in turn, to the still controversial point about the agency of objects. Whether we praise objects for their agency or brashly deny that they have any, we overlook the question of what objects are when not acting. To treat objects solely as actors forgets that a thing acts because it exists rather than existing because it acts. Objects are sleeping giants holding their forces in reserve, and do not unleash all their energies at once.

Since it cannot be assumed that readers of the present book are deeply familiar with OOO, it will now be necessary to repeat some points already known to readers of my previous books. Enough time will remain afterward to add new twists capable of surprising even the most grizzled OOO veteran.

2.
The Dangers of Duomining

There are only two basic kinds of knowledge about things: we can explain what they are made of, or explain what they do. The inevitable price of such knowledge is that we substitute a loose paraphrase of the thing for the thing itself. Whether we speak of a poem, a corporation, a proton, or a mailbox, something is changed when we try to replace an object with an account of its components or its effects, as literary critics have long known. (See Brooks 1947.) In technical terms, the attempt to paraphrase objects always amounts to undermining, overmining, or duomining them (Harman 2013).

An object is undermined if we explain it in terms of its smaller constituents, by way of a downward reduction. Western science was born in undermining, when the pre-Socratic thinkers of ancient Greece aspired to find the ultimate root that explained the composition of mid- and large-sized entities. Are all extant things made of water, air, fire, atoms, number, a formless lump, or something else altogether? Undermining has remained the dominant method of physics but is less common in social theory, which does not work in the idiom of ultimate particles. A surprising counter-example is Tarde, who bases his sociology on tiny monadic substances that form larger beings only by grouping together under a single dominant monad, rather than by forming a larger compound entity per se (Harman 2012a). Undermining can also occur when authors emphasize the dependence of humans on their background conditions, as in the landmark New Materialist anthology of Diana Coole and Samantha Frost: “Our existence depends from one moment to the next on myriad micro-organisms and diverse higher species, on our own hazily understood bodily and cellular reactions and on pitiless cosmic motions, on the material artifacts and natural stuff that populate our environment” (Coole & Frost 2010, 1). The problem with undermining is that it cannot account for the relative independence of objects from their constituent pieces or histories, a phenomenon better known as emergence. An object is not equal to the exact placement of its atoms, since within certain limits these atoms can be replaced, removed, or shifted without changing the object as a whole. Nor is an object identical with the influences received from its environment, since some of these remain ineffectual while others prove decisive. Rome, Athens, and Istanbul might be the same cities as in ancient times despite complete population turnover and radical cultural and infrastructural change. An object is more than its components, and hence cannot be paraphrased successfully by way of downward reduction.

But the greater danger for the humanities and social sciences is the opposite one, overmining. Here, rather than treating objects as superficial compared with their ultimate tiniest pieces, one treats them as needlessly deep or spooky hypotheses by comparison with their tangible properties or effects. Eighteenth-century empiricism tells us that the object is nothing but a bundle of qualities; contemporary thinkers say instead that the object is nothing but its relations or discernible actions. Latour is surely the most stimulating present-day thinker of overmining, as in his daring claim that “there is no other way to define an actor than through its actions, and there is no other way to define an action but by asking what other actors are modified, transformed, perturbed, or created” (Latour 1999b, 122). The problem with overmining is that it allows objects no surplus of reality beyond whatever they modify, transform, perturb, or create. In this way ANT unknowingly repeats the argument of the ancient Megarians, who claimed that no one is a house-builder unless they are currently building a house, a claim refuted by Aristotle in the Metaphysics (Aristotle 1999, Book Theta, Chapter 6). For if objects were nothing more than their current expression in the world, they could not do anything differently in the time that follows. No “feedback loop” can replace the need for an excess in things beyond their relations, since an object cannot absorb or respond to feedback unless it is receptive, and this requires that it be more than what it currently does. Just as we saw that an object is more than its components, we now see that it is less than its current actions. The author Harman who currently types these words in the University of Florida Library while wearing a black sweater is far too specific to be the Harman who will leave Florida next Sunday and can remove the sweater whenever he pleases.

It is rare to find undermining or overmining strategies in isolation. Usually they combine in mutual reinforcement, in a two-faced reduction known as duomining (Harman 2013). The earliest duominer in the West was Parmenides, who proclaimed a double cosmos with a single unified Being on one side and a truthless play of opinion and appearance on the other. Everything was either pure unified depth or pure variegated surface, with no intermediate room for genuine individual things. Another example is found in certain forms of scientific materialism, which ruthlessly undermi...

Table of contents

• Cover
• Title Page
• Copyright
• Part One: Immaterialism
• Part Two: The Dutch East India Company
• References
• End User License Agreement