What I have to say today about intentionality builds on a simpler conception of intentionality that I developed previously. Thus allow me to say a few words about it before I proceed. First I would claim that properly speaking we are not intentionally directed at objects in the world at all. Rather we are intentionally directed at object possibilities. An object possibility we can understand as a range of objects, bounded by some constraints. For example, if we have seen only the front of an object then the object possibility we are intentionally directed at is a range of possible objects all sharing that front, but having any reasonable backside (one we wouldn’t be shocked by). If this object possibility “fits” a real object to some degree then we might say that we are intentionally directed at that object, but that is a third person way of speaking. If the object possibility we were directed at did not “fit” any real objects from our point of the intentional directedness is the same, it is only from the third person point of view that someone might be inclined to say that we weren’t really intentionally directed at anything at all. And, finally, allow me to say what these object possibilities are defined in terms of. In the vast majority of cases they seem to be defined in terms of sensory input, and how that input would change in response to our interactions with the object or the object being affected by other forces. But we can allow for the possibility of intentionality directed at abstract entities, such as numbers, by allowing the input to also include ideas or concepts, the object possibility being defined by that input and how it changes as a result of abstract operations upon it.
This model of intentionality implies that each object possibility is defined over basically a single domain. A tree, for example, seems like it would be defined entirely in terms of sensory inputs. But consider what happens when you unknowingly encounter a fake tree. You believe it to be real, but regardless of whether it is an actual tree or not you are intentionally directed at it through the kinds of sensory impressions you can have about it. But one day you find out that the tree is not really a tree at all, but a fake. With this revelation something, something hard to define, happens. You see your previous intentional directedness at the fake tree as in error somehow. Although you realize that you were intentionally directed at the fake previously you will not be intentionally directed at in the same way in the future. Let me provide another example of this phenomenon at work. This time the tree is well, but one day it is uprooted and dragged somewhere else. As you walk by it you are intentionality directed at it, but as some unknown uprooted tree beside the road. But if someone tells you this was the tree that you had previously encountered you will again experience the way you are intentionally directed at it change. Your previous intentional directedness at it, as an unknown uprooted tree, will again be seen as in error somehow, although it was still directed at it, just not in the “right” way.
Similar phenomena can also occur in language and perception, which is no surprise since both reduce to intentionality in significant ways (meaning that explaining either requires some appeal to intentionality). In language, for example, we have the famous example of the martini drinking man, who is really drinking something else (and may not even be a man), but who appears that way to you. You can communicate successfully about him using those terms, even if your listeners know that he is not drinking a martini (probably). However, if you become aware that he isn’t really drinking a martini it will become apparent to you that somehow your intentional directedness at him was flawed. But in that case how were you successfully able to communicate? Similarly, in perception, you may experience a vivid hallucination of a purple rat. Your perceptions about this rat are caused by some internal glitch. Thus third parties may agree that what your thoughts about the purple rat are really about is the glitch. However, in terms of your intentional directedness at the purple rat, it is clear that if the rat doesn’t really exist then your intentional directedness is a failed directedness, one with no actual objects, from your perspective.
All of these cases can be explained in basically the same way, or so I claim. To account for them I would revise my original explanation of intentionality. In my original explanation I defined the object possibilities as being defined by a particular kind of input, for example sensory input. To this I would add that there are a number of “layers” to this possibility. Each layer is defined by a different kind of input. For example, in the case of our tree there are three layers that seem relevant. One is the perceptual layer, which defines the tree as having a particular appearance. Another is a conceptual layer, which identifies it as a tree, a concept that is tied to other concepts and perceptual inputs in ways that define it. And we might add to the conceptual layer other abstract properties we associate with the tree, such as location, or we might relegate them to their own layers. Finally we have the object layer, which defines the tree as a single object existing over time.
Not all of these layers are equally important. What we consider our own intentionality as actually being directed at, after we learn all the facts, depends on which layer or layers are the strongest. I am not going to commit myself to a claim about which layers are stronger and why at this time, but I will use the idea to explain the examples provided previously. Let us turn to those examples then. In the case where the tree turns out to be a fake the perceptual layer and the object layer hold, they are still directed at the fake tree, although the conceptual layer no longer matches up with it. However, the conceptual layer is weaker, and so we “fix” our intentionality by revising it. The fact that it doesn’t match the other layers is what causes the feeling that something has gone wrong. And the case with the uprooted tree is similar, although in this case it is the perceptual layer that is out of line with the others. The sensation that something has gone wrong intentionally is, in this case, caused by our revising of the perceptual layer and conceptual layers with which we were previously directed at the uprooted tree with a new object layer. In the case of the martini drinking man our language conveys the conceptual level, that the object of our intentions is a man and is drinking a martini. But our listeners, if they cannot find anyone to match to that conceptual level (and thus add a perceptual layer of intentional directedness of their own as well), will reconstruct our perceptual layer (looking for someone who looks like a martini drinking man) and intentionally directing themselves at that person. Finally, in the case of the purple rats our perceptual layer is directed at the glitch, because it is the source of those perceptions, but the object and conceptual layers fail to get a grip on anything, and so we consider our own intentional directness failed and without any real object, although there was something that satisfied part of it.