1999-07-22

Carolyn Ray [Ray99] sets up the foundations of identity theory from a conceptualist perspective and solves most of the traditional problems and puzzles of modern identity theory. Ray's work is pa rticularly strong in its analysis of personal identity and the central role this plays in all areas of identity theory. However, while making substantial progress in the area of the identity of non-persons, Ray makes an error in her analysis of actions an d the meaning of reified actions, to the extent of arguing that some perfectly ordinary human behaviour -- referring to fictional characters as if they were entities -- is a mistake. In this paper I start by describing what bothers me about Ray's conclusi ons and then get to an analysis of her very clearly stated premises.

My argument is that actions can produce edges, and inside these edges are unities that we can identify as entities, and these are the entities that we refer to when we reify actions. This analysis is carried out almost entirely within the framework th at Ray lays out, and should be understood as an elaboration upon that framework rather than a fundamental criticism of it.

With regard to "Elizabeth Bennet, the well-defined principle figment in Pride and Prejudice", Ray [Ray99:Ch8] writes:

Although I am aware that focusing on the normative value of the ordinary can lead one into all kinds of error -- from Aristotelian physics to conservative politics -- it is still important to examine carefully cases where philosophical conclusio ns differ radically from ordinary experience. To see why we should want to examine Ray's claim more carefully it is worth asking what in reality MISTAKE refers to.

A MISTAKE is a type of action. In particular, a MISTAKE is an action that has different -- usually negative -- consequences from those intended. If I intend to hit a nail with a hammer and hit my thumb instead, I have made a mistake: the consequence of swinging the hammer is a sore thumb rather than a driven nail. By this definition, is talking about fictional characters as if they are entities a mistake? What are the consequences of doing so that we did not expect? And are they negative consequences? Or can there be mistakes that don't have negative consequences?

Looking more deeply, it is clear that sometimes we make a mistake that has no effect on the final outcome of an action, but that does have an effect on the intermediate outcomes. If I am doing some simple arithmetic, I might write: (2 + 4) + 7 = 5 + 7 = 13. Now, clearly I have made not just one mistake but two, and reaching the intended outcome -- the correct answer -- is a matter of accident.

I believe this analysis of MISTAKE is sufficient for my purpose here. If it is a mistake to talk about Elizabeth Bennet as if the name referred to an entity, then either there are consequences that are unintended, or we reach the intended consequences by accident.

So to understand if we are making a mistake, we must understand our intent in taking about Elizabeth Bennet as if there was an entity of that name, and then see if we wind up achieving something quite different from our intent, or if we achieve our int ent only by accident. Ray's account of our intent is: "What we are really trying to understand is what Austen intended her figment to represent." It is true that in this case we are probably making a mistake. Thinking about Austen under the pretense that we are thinking about an entity named Elizabeth Bennet is certainly capable of impeding our understanding of Austen's intent.

It is not true, however, that everyone talking about Elizabeth Bennet has the same intent. I certainly don't, at least not at the moment: just like Ray, I am talking about her to understand the nature of objects of the imagination, and to motivate an a rgument that justifies reifying certain types of action -- including the mental entities we call concepts -- and understanding how we can thereby properly consider them entities. But while I write I am acutely aware of Elizabeth meeting Mr. Darcy in the p ark, and she seems quite real to me, and I cannot detect any unintended consequences that would indicate I am mistaken about this.

Could I be making a mistake and fail to realize it? Of course. But now that I have had the possibility of a mistake pointed out to me, why is it that I can't imagine what the negative consequences of my mistake could be? In contrast, consider Gilgam esh. I happen to think that there probably was a man who "built the mighty walls of Uruk of the sheep-fold," and who might well have been possessed with the desire for immortality after his friend Enkidu died. I enjoy the poetry even more because I bel ieve this. If I were proven to be mistaken, if it turned out that the archeological record showed that Uruk was never walled, or was always under the suzerainty of foreign kings, or the like, I would feel mildly disappointed because I had made a mistake: I had failed to understand reality and had instead believed a falsehood.

But in the case of characters that we know are the product of someone's imagination, there do not appear to be any unintended consequences of talking and thinking about them as if they were entities. Having been told that I am making a mistake when I t alk about Elizabeth Bennet as if she was an entity, I am at a loss to detect any unintended consequences of my thinking, and so I am doubtful that I am making a mistake at all. If I were really making a mistake about the reality of Elizabeth Bennet then I would be properly considered delusional, and I am not. On the contrary, talking about Elizabeth Bennet and other figments is an enjoyable and positive part of my life, and I find it implausible in the extreme that a mistake is responsible for this value.

This is the core motivation in my present disagreement with Ray: I do not believe that value can be derived routinely and reliably from a mistake.

So: granted that Elizabeth Bennet and her ilk are entities, what kind of entity are they? And why does Ray insist that they are not entities at all?

Ray contends [Ray99:Ch8] that all reification of actions is in error:

It follows quite naturally from this that:

And therefore it is a mistake to say that there is an entity, "Elizabeth Bennet" -- not just that there is no woman named Elizabeth Bennet, but that there is anything at all with that name. The claim that we fail to refer in these cases is very strong, and it is my intent to show not only that we refer, but to show what we refer to.

In her analysis of UNITY Ray discusses EDGES [Ray99:Ch6] although she does not define the concept:

Because UNITY and IDENTITY are different perspectives on the same fact about entities, it is worth looking at UNITY to see if we can understand how an action can be intimately associated with a unity -- so intimately associated that were the action to cease, the unity would no longer exist, and the entity that is that unity -- that has that identity -- would also no longer exist.

To understand UNITY we really need to understand EDGE. What in reality does EDGE refer to? Human beings are relentlessly spacio-temporal in our orientation, and our concept of EDGE is formed by noticing that the juxtaposition on one dimension of very d ifferent values along another dimension.. Typically the former dimension is space or time, but I do not believe it has to be. The latter dimension may be density, colour, or any other attribute.

Ray gives a good deal of attention to density, calling density boundaries "metaphysically real" as opposed to, say, colour boundaries. But clearly all edges that are the result of changes in any property that is independent of the observer are "metaph ysically real", for this is just what metaphysically real means: independent of the subject. Colour boundaries are less interesting than density boundaries because they do not bound anything in three dimensions unless there is a density boundary that go es along with them. Thus the strange example that Ray cites of a people who use colour rather than density to identify entities fails simply because the purported colour boundaries don't actually bound anything.

Natural language -- at least English -- is apparently a conspiracy by naïve realists to impose their views on the rest of us. I use "metaphysically real" strictly in the sense of "not a property of the relation between the subject and object, but of the thing." I use "thing" to mean the thing independently of subject, and "object" to mean the thing including properties of its relationship to the subject. In all of this remember: the subject is metaphysically real as well; the subject has prop erties that are independent of its relationship to any other subject. And the relationship between the subject and the object is also metaphysically real, and can be referred to as such if it is the relationship rather than the thing that is the object o f our attention.

In this sense, there are edges that are not metaphysically real: they are due to pure acts of attention. Suppose I am an aircraft navigator: I may well consider the volume of air outside my aircraft to be an entity. To be concrete, I may consider a block of air 100 m on a side. There is no metaphysical boundary in this case. No one other than me would ever isolate such a volume: there is nothing about it other than its relationship to me on which to base such isolation. There is no sudden chang e in any metaphysical attribute of the air, but rather a sudden change in my intent as my attention crosses the boundary. This still fulfills the definition of EDGE as a sudden change along one dimension as our attention moves along another dimension. T he block of air inside the edge is a unity, and the unity is an entity.

An ENTITY is some part of reality abstracted from its surroundings by an act of selective attention. As in the example above, the basis for abstraction does not have to be a metaphysically real property, but may be an objective property: that is, a p roperty of the relationship between me and the thing. Entities, in Ray's account and my own, are what have identity.

An entity may have internal edges as well; depending on our purpose, we may count one or many entities. My children play with Lego a good deal, and when one of them builds a ship and shows it to me, I can see either a single entity (the ship) or a wh ole bunch of entities (the individual Lego blocks) depending on what edges I attend to. Sometimes the structure has a particularly weak join somewhere, and in this case I can consider, say, "the front half of the ship" as one entity and the back half as another. If the structure is changed to eliminate this weak join, these entities go out of exist metaphysically, although they may still exist objectively.

As Ray insists, actions are actions of entities. But this is not the whole story.

Consider two cases of entities acting. In the first case, the actor is a philosopher, Carolyn Ray. The action is running. Consider the scene carefully: arms and legs move smoothly in perfect rhythm, the action is inseparable from the actor, and the actor is not sub-divided by the action. This last is the crucial characteristic of this kind of action: it does not create any edges.

In the second case, the actor is the Pacific Ocean. There is a physicist floating in it beside a reef, in SCUBA gear. The ocean surface is flat, but he notices a disturbance approaching. He is lifted up and flipped over the rock. The action is wavi ng. In this case, the action is inseparable from the actor, but the actor is sub-divided by the action. The waving creates an edge where there was no edge before. Inside the edge is a unity, an entity: a wave.

In this limited sense we can justifiably reify actions, and our use of entity-terminology in these cases is not a metaphor or a mistake but literally true. Real entities being created by real actions: the entity and the action are completely co-spatial and co-temporal. As soon as the action ceases, the edge -- and the entity -- cease to exist. The entities are created and sustained by the active, ongoing, creation of edges in previously homogenous media. In the case of waves this should no t be too controversial. But what about the case of thoughts?

Is a concept an entity? I hold that it is in exactly the same sense that a wave is. My own internal experience -- my consciousness -- is pretty undifferentiated: when I am not deliberately thinking about something in particular I am not aware of any edges in my awareness, in my internal state. When I do think about something in particular, when I focus and concentrate on a subject, I am aware of edges created by that mental activity. These edges are created by the actions of my mind necessary for the apprehension of concepts and acts of the imagination. Ray is correct in saying, "Concepts do not persist; I persist. My concepts recur." But the action of their recurrence causes edges in my mind, and inside the edges are the entities I refer to when talking about the concepts.

As with concepts, so it is with objects of the imagination. There is an entity, created by the action of my mind, named "Elizabeth Bennet." This entity has properties that are very similar to the properties of a memory, but instead of being created b y my perceptual experience of a living woman, it was created by my reading a book. Anything that I might remember about a woman I might also know about Elizabeth Bennet, and if you and I have read the same book the entities we created in our minds are li kely to be similar; so similar that we may be able to treat them as identical. But they are not identical: there are at least as many Elizabeth Bennets as there are readers of Pride and Prejudice. This is one of the things that is likely to mak e our conversation interesting. It is as if we both owned sheepdogs, but not the same sheepdog. We can sit down and talk about how our dogs behave, how defensive they are and that funny way they lower their heads to get a good look before they bite some one.

Thus, Elizabeth Bennet is in fact a concept that subsumes all of the mental entities named Elizabeth Bennet in the minds of all the readers of Jane Austen's novel.

My purpose here has been to show that -- contra Ray -- we do refer when we talk about figments, and in general we can reify actions that create edges in previous homogenous media. My argument was motivated by the belief that mistakes ca nnot routinely create value, and guided by many years of experience with waves of various types.

All edges are equal epistemologically, but some edges are inherent or metaphysical while others are contextual or relational. When we form the concept EDGE one of the measurements we omit is the degree of contextuality in a given edge. Although it di d not directly impact the present argument, thinking about this aspect of edges was a considerable aid.

I view this argument as a bit of polishing on the framework that Ray has established, rather than anything that runs directly contrary to it. The basic structure of her analysis of identity and the puzzles of identity theory are sound, and should form the basis for further elaboration of the nature of identity. In particular, an important concept that Ray uses but does not analyze in detail is: REPLICA, which may be a subject of future work.

Ray99: Carolyn Ray, Identity And Universals: A Conceptualist Approach to Logical, Metaphysical, and Epistemological Problems of Contemporary Identity Theory, Ph.D. Thesis, Indiana University, 1999. <back>