Wednesday, July 18, 2012

From Objects to Pluralism

I have now revised this and republished it HERE. The major change is to incorporate Harman's notion of vicarious or indirect causality as the basis on which to examine patterns of relations among objects.

* * *
. . . we are not to attempt to hack off parts like a clumsy butcher, but to take example from our two recent speeches. The single general form which they postulated was irrationality; next on the analogy of a single natural body with its pairs of like-named members, right arm or leg, as we say, and left, they conceived of madness as a single objective form existing in human beings. Wherefore the first speech divided off a part on the left, and continued to make divisions . . . 

– Plato, Phaedrus (265e-266a)
Having hazarded that pluralism is the Next Big Thing I now feel some obligation to clarify what I mean by pluralism. As it’s object-oriented philosophy that brought me to this dance, I’ll use it as a vehicle for so doing.

First, using a passage from a Graham Harman interview, I raise the question of the relationship between philosophy and the more specialized disciplines. I then continue with Harman in a section where, in effect, I ask: What can we build with objects and relations alone? By way of illustration I bring up the case of knowledge representation in the cognitive sciences, where complex conceptual systems are constructed from just that, objects and relations.

Then I take an excursion into the work of Levi Bryant, whose concept of regimes of attraction indicates the existence of relatively stable patterns of relationships over large collections of objects. I then go into full tap dance mode, suggesting that we can construct Realms of Being from that notion. Realms of Being, that the world consists of many different ever evolving Realms, THAT’s what I mean by pluralism. Given that, the task of metaphysics is to figure out what those Realms are and how they’re interlinked.

I conclude with some more general remarks.

A General Theory of Objects?

As a way of setting the stage, consider the following passage from Graham Harman’s interview at ASK/TELL:
. . . the reason to focus on objects rather than on “language, social change, sexuality or animals” is because philosophy is obliged to be global in scope. If philosophy were to give one of these other entities a starring role, it would have to reduce the rest of the universe to them. “Language is the root of everything.” Here, you are choosing one specific kind of entity to be the root of all others, and there is no basis for this. Sociology tends to view all reality in terms of its emergence from human societies and belief-systems. Psychology treats all reality as made up primarily of mental phenomena. Physics deals with tiny physical objects and says that everything is made out of them, except that physics is useless when trying to explain things like metaphors, the Italian Renaissance, the meaning of dreams, and so forth.

All these other disciplines focus on one kind of object as the root of all else in the world. Only philosophy can be a general theory of objects, describing Symbolist poetry and the interaction of cartoon characters just as easily as the slamming together of two comets in distant space.
My immediate and quite spontaneous reaction to that was a less than charitable: And just what can philosophy tell me about cartoon characters? I asked that question in my capacity as someone who has a specialized interest in cartoons and so has spent hours upon hours going through cartoons scene by scene, shot by shot, and even frame by frame, trying to figure out how these things work. It would be too much to expect a philosopher to look at cartoons in such detail.

But just what WOULD I expect of a philosopher? I don’t need a philosopher to tell me that Popeye is, in some sense, real. I know that already, that’s why I care about them and study them. Nor do I need a philosopher to tell me about the difference between the real object on paper or in celluloid and the image in someone’s mind. That’s been around for a long time. I don’t see that philosophy has anything new and interesting to say about that.

But then, just what does philosophy have to offer the other specialized disciplines? Do they have need of Harman’s “general theory of objects”? I have my doubts. Does ANYONE, other than philosophers, have need of a general theory of objects? If the answer to that question is “no” does that mean that such a theory has no use?

We are now in very dangerous territory. I want to make one not-so-digressive remark and then continue on by suggesting that perhaps what philosophy has on offer, even Harman himself, is not quite or not merely a general theory of objects, but something nearby.

That not-so-digressive remark is that literary criticism has, in the past few decades, drawn on various philosophies as interpretive systems. I figure that any half-way interesting philosophy can serve in that capacity and serve, not only literary criticism, but cultural criticism and related humanistic pursuits. Indeed, these disciplines tend to shade into philosophy, at least philosophy of the Continental kind, if not the Anglo-American kind.

But Harman’s claim, it seems to me, is wider. He’s claiming, if I read him rightly, all of knowledge and not just the hermeneutical humanities. Philosophy, once again, is going to cover the whole territory. I think the claim is a valid one.

I also believe we need to think, not just about objects, but about what I’m provisionally calling Realms of Being. Specifically, what I propose is that we use a general theory of objects in explicitly constructing an account of Realms of Being where the Realms constitute an unfolding large-scale organization of objects.

Building Materials: Objects and Relations

Let us continue with Harman, who has said that his philosophy has two basic principles:
1. Individual entities of various different scales (not just tiny quarks and electrons) are the ultimate stuff of the cosmos.

2. These entities are never exhausted by any of their relations or even by their sum of all possible relations. Objects withdraw from relation.
Yes. And furthermore, in The Quadruple Object Harman asserts that “the basic rift in the cosmos lies between objects and relations in general: between their autonomous reality outside all relation, and their caricatured form in the sensual life of other objects” (pp. 119-120).

What I propose is that that’s all we need to construct an account of the Realms of Being, that and a university’s worth of knowledge from the specialized disciplines to inform the construction. Philosophy cannot exist in a vacuum.

What do I have in mind? Something a bit like mathematics. Mathematicians build rich conceptual structures using only a few foundational notions. But mathematics is not my immediate model.

Knowledge representation is. Knowledge representation arose in the cognitive sciences in the 1970s and 1980s where methods were needed to represent human knowledge in computers. Some investigators were interested in solving practical problems, e.g. so-called expert systems for medical diagnosis or the configuration of large computer systems, while others were interested in simulating human reasoning. In all cases researchers needed a way of representing knowledge in a form that was tractable by computers but also rich, flexible, and robust.

Many schemes made us of graph theory, a branch of topology. A graph is a network:


In artificial neural networks the nodes in the graph represent neurons while the links between the nodes represent connections between neurons. In cognitive or semantic networks the nodes typically represent concepts while the links between nodes represent relations between concepts. Here’s a simple example:

This network, while it has fewer elements than the first one—and both are just very small fragments of real networks—is nonetheless more complicated because it has more information. The nodes and links are all labeled and those labels are important.

Notice that we have two kinds of nodes, square and round. The round nodes are entities while the square ones, of which there is only one, are actions. Person, Fred, hit, bat, ball, and thing are all concepts.

Correlatively, ISA (is-a), AGT (agent), STR (instrument) and OBJ (object) are relations between concepts. ISA means, well, is a: Fred is a person: bat and ball are things. Fred is an agent (AGT) in an act of hitting where bat is an instrument (STR) and ball is the object (OBJ)—where object is understood in a cognitive or linguistic sense, not a metaphysical one.

So, in this little fragment we’ve got two kinds of concepts, entities and actions, and four kinds of relations between them, agent, instrument, object and is-a. This, of course, is only the tiniest fragment of a real cognitive network. And real networks have more than two kinds of concepts and four relations. Maybe we’ve got a dozen or more kinds of relations and a similar number of conceptual kinds. But not much more, as the idea is to represent as much knowledge as you can using as few primitive elements (concepts and relations) as you can.

I’m proposing something similar in metaphysics. Harman says that the world consists of objects, on the one hand, and relations among them on other. OK. So what can we build from that? I understand quite clearly that constructing a representation of how people think about the world is different from constructing an account of the world itself. One justifies a representation of thought by invoking evidence about how people actually think. One justifies a representation of the world by invoking evidence about how the world works, evidence that is the province of many specialized disciplines. Physicists, geologists, chemists, art historians, and so forth don’t operate as though each object under their purview is utterly unique in kind nor that each relationship between two or more objects is utterly unique. They group objects into classes of different kinds.

That’s all we need. Let’s go back to the passage I’ve quoted from The Quadruple Object, this time continuing on (pp. 119-120):
...the basic rift in the cosmos lies between objects and relations in general: between their autonomous reality outside all relation, and their caricatured form in the sensual life of other objects. Whatever the special features of plants, fungi, animals, and humans may be, they are simply complex forms of the gap between objects and relations, just as heavier chemical elements arise from hydrogen and helium. By no means does this imply that mentality is reducible to neuroscience or string physics. For our principles forbid that any specific kind of entity could be the building block for everything else in the cosmos. Instead, everything plays out in the strife between concealed objects and the twisted of translated forms in which they appear to other objects.
I’m proposing only that those “complex forms of the gap between objects and relations” have a pattern and I’m calling that pattern Realms of Being.

What is more I believe that one object-oriented philosopher, Levi Bryant, has already taken a first step along this path.

Patterns of Relations Among Objects

Let’s look at a passage Bryant posted on 27 October 2011 in Do Attractors do Anything?:
With Graham and Latour, objects have to go through all the translations and transformations to get from one object to another. In other words, for me acorns do not virtually contain oak trees. Rather all sorts of translations have to take place to get from acorns to oak trees, and the oak tree that evolves from the acorn is a genuine and novel creation in the universe. There’s nothing that is pulling the acorn to the oak tree. Potentiality or virtuality are important dimensions of objects for me (and here I guess Graham and I still diverge), but when I think of virtuality/potentiality, I don’t have something like the acorn containing an oak tree in mind, but rather something more like the potential energy contained within a tautly drawn spring or rubber band. The translations still need to take place. The translations still need to take place. Virtuality also just means that something must be susceptible to affecting and being affected by other things for interactions to take place.
As Bryant well knows, acorns cannot become oaks all by themselves. There is no little oak inside the acorn (as the preformationists believed) that just gets bigger and bigger. The acorn requires certain causal forces, or opportunities, if you will, in order for an oak tree to sprout and grow.

And that implies a problem. For the point of objects are that they are autonomous. But in what sense can an oak tree be autonomous if its very existence is dependent upon a proper environment? Bryant is aware of the issue, though I’m not quite sure that he resolves it. Here’s a passage from The Democracy of Ideas, Chapter 5 (p. 196):
In this connection, we can ask ourselves how it is possible for objects to be constrained despite their autonomy, independence, and self-determination. In many respects, it is the distinction between virtual proper being and local manifestation, coupled with the concept of regimes of attraction that allows us to theorize these constraints. For while, in their virtual proper being, objects withdraw from any of their actualizations in local manifestations, while every object always contains a reserve excess over and above its local manifestations, nonetheless local manifestations are often highly constrained by the exo-relations an object enters into with other objects in a regime of attraction.
The language of constraint allows him to conceive of the object as autonomous. That is, the object, in its virtual proper being, really is free and independent it’s just that, um, err, well, in any actual context it is subject to constraints. I submit that the on the obverse of at least some of those constraints we find dependency, resources on which the object depends. The acorn will not become an oak without water, sunlight, nutrients of many kinds, who knows what microbes in the soil, soil of a proper consistency to hold the root system, and so forth. At the same time, the availability of these things will constrain the growth of the tree.

So, there IS a problem, which I wish to set aside for a moment. What's important is simply the acknowledgement of all those necessary exo-relations. Bryant has a term for them in Democracy, Chapter 4 (pp. 169-170):
I refer to networks of exo-relations like this as “regimes of attraction”. Regimes of attraction are networks of fairly stable exo-relations among objects that tend to produce stable and repetitive local manifestations among the objects within the regime of attraction. Within a regime of attraction, causal relations can be bi-directional or symmetrical or unidirectional or asymmetrical. Bi-directional causation is a circular relation in which two or more entities reciprocally perturb one another in response to each other.
That’s what I’m after: “networks of fairly stable exo-relations that tend to produce stable and repetitive local manifestations among ... objects.” A regime of attraction* is a pattern over objects and relations. All oak trees require pretty much the same regime. Details will vary, of course, but the general requirements are fixed.

My proposal is that, when consisted at a sufficiently high level of generality and ranging over a diverse collection of objects, a regime becomes, in effect, a Realm of Being. Living Things could thus be thought of as a Realm of Being. All living things require and are dependent upon energy input and appropriate nutrients. That common dependence, I suggest, binds living things together into a Realm which is ontologically separate from the Realm of inanimate things—I say “ontologically separate” because, obviously, living things must be physically intermingled with nonliving things. Within the Realm of living things there is perhaps a distinction between a Realm of single-celled beings and a Realm of multi-celled beings. The Realm of living things may contain other Realms as well—almost certainly—but there are Realms of inanimate objects and human life has realms of its own that are peculiar to it, realms mediated by symbol systems.

I have no concrete list of Realms to propose. I’m not anywhere near such a thing. I’m just offering a general idea.

The important point is just that objects are organized into, organize themselves into, Realms of Being according to patterns of relations among them.

Basics Revisited

With that in mind let me edit and amend Harman’s statement of the basic requirements of his metaphysics. The first two statements below are edited from Harman while I’ve added a third:
1. Individual entities of various different scales are the ultimate stuff of the cosmos.

2. These entities enter into relations with other entities but are never exhausted by any of their relations or even by their sum of all possible relations.

3. Realms of Being consist of specific kinds of entities in specific relations with one another.
I will even hazard a fourth proposition:
4. Our cosmos has evolved from one Realm to the many evident today. It is possible that Realms exist of which we are unaware. There is no obvious limit to the emergence of new Realms from existing ones.
Notice that I have removed Harman’s language of withdrawal. I prefer a language of plenitude, abundance, or fecundity.

It is because objects are inherently fecund that they can enter into an unbounded number and pattern of relations without exhaustion. Some of those patterns coalesce into new objects established upon new patterns of relations among themselves, thus yielding a new Realm with its own laws and relations. While the objects of this new Realm are constructed of objects and relations in pre-existing realms, they cannot be reduced to those objects and their relations.

So, What is the Scope of Metaphysics?

The scope of metaphysics thus DOES range over all the disciplines, as Harman asserted. And a general theory of objects is critical to this endeavor. But we need to augment that theory with a conception of patterns over objects and relations that are so consistent and widespread as to constitute Realms of Being.

It is the job of the metaphysician to identify those Realms. To do this the metaphysician must needs consult with in the many specialized disciplines, not to critique nor to reconstruct, but learn. What must the metaphysician learn? Whatever is necessary to get the job done.

That will have to be negotiated, negotiated among metaphysicians and specialists, and among metaphysicians themselves. I see no way of setting guidelines before the fact. The only thing to do is to wade in and get muddy with details.

* * * * *

* The term itself, “regime of attraction,” is an unfortunate example of the scientism that infests Bryant’s thought, but I don’t want to argue that point here as I’ve already done so in two other posts, one on entropy and another on attractors and phase space. The important point is the Bryant’s insight, that we should examine “networks of fairly stable exo-relations among objects”, does not depend on his misconstrued notions of phase space and attractors.


  1. Hello Bill,
    I am of course sympathetic to the program that you announce in the title, and to your notion of multiple realms. I am glad that you agree that the language of withdrawal is a bad idea, and that you espouse the language of fecundity and abundance. The only points of divergence I have are
    1) "objects" seems to me to pre-decide on the nature of the elements of the world, it is too definite a term for me. While I am in favour of the distinction between elements and relations, i do not make a universal dogma of it. In some cases or in some realms the terms may be relational all the way down. Only empirical research can decide.
    2) your realms seem to imply a stratified model, of one realm building on the objects in the previous realm. Feyerabend allows manifest realities to exist side by side, without one realm emerging out of another eg scientific physics and Homeric gods.
    Feyerabend wanted an ontology that left as much as possible to empirical research, and that protected as much as possible the abundance and plurality of reality.

  2. Hi Terry,

    On both the issues you mention I'm willing to let empirical evidence decide.

    Specifically, about relations, I don't think we're going to get far simply thinking in terms of relations as though all relations are the same. They're not. But what kinds of relations do we need? That's not obvious to me.

    For example, we know that birds navigate by the stars (as, of course, do we). What kind of relation is that?

    As for stratification, certainly there's going to be stratification. But I don't think that's how it will always be. For example, I know biologists have been talking about 'major transitions' in evolution.

    One such transition is from single-cell protists to multi-cell creatures, such as plants, animals, fungi. The emergence of multi-cell organisms from protists would be stratified, with the multi-celled 'above' the single cell. But what of the distinction between plants and animals? Are they two different Realms? Off hand, I have no opinion and, in any event, I don't know enough biology for my opinion to be worth much. But I can easily imagine that, yes, they're two Realms, but two Realms in parallel.

    And when it comes to the Realms humans have created through culture, I expect a lot of parallelism there. And, after all, that's where my interest lies. If different (scientific) paradigms—and we’ll need to be very careful in how we identify them—qualify as different Realms, then we’re going to have a blither of them. And, who knows, we might need real honest-to-god topology to chart their interrelations.