Circularity is a central charge some philosophers make against inner theories of meaning, where we are said to essentially do an inner lookup against a mental filing cabinet to understand a word, a diagram, and so on. The problem is that the inner lookup is offered as an explanation for how we understand something, but it must presuppose the understanding that it sets out to explain.


Note that I don't bother with the “how” in regards to understanding and the like. That's what I mean by not switching to a “wires and pulleys” explanation. Discovering the “how” will involve more than just drawing an analogy to a person retrieving a file form a filing-cabinet. In lieu of that we can note the types of situations we use the word “understand” in and for what purposes. In that broader context we can see that understanding is a form of success in life.



You would have me ask the student what his model is. He tells me. I complain to you, “Those are just words. Where is the model?” It's clear that I will never find out what the student's model is because all he can do is say things to me or show me things, all of which you suggest are not the model. The model is supposed to provide understanding. I'm trying to drill down to the entity that you think provides understanding, unification, and all the other selling points you have made about models.



We both need to be careful. One might validly charge here that this is being made up as we go along. There is no process of retrieval-by-address that we witness. We would be saying what must be true for the theory to hold. The claim is now that we can match labels to addresses all prior to having any model in hand. Consider if the “label” is a brand of apple that you haven't seen before, but that you identify anyway as an apple. There would be no address for that specific apple, yet you knew which model file to retrieve from the cabinet. The unification the model was supposed to provide occurred before you had the model in hand.



Let's consider me drawing that circle again, botching it, erasing part of it, and correcting it. You say that I must be comparing to an inner circle to be able to correct the circle. But from the perspective of my eyes relative to the circle on the paper, I don't see a circle, but a geometrically projected circle. In fact my head and eyes are probably in constant motion, meaning I am viewing a changing figure on the page. I could not be comparing to an inner perfect circle.


Yet, after correction at least, I recognize the figure on the page as a circle. You might say that I must have built a model of a circle that incorporated all projections, but consider that even a child recognizes a circle from various perspectives, yet is in no position to master projective geometry well enough to build a model that incorporates it. If I ask him what his model is, he surely won't be telling me about a principal axis and transformations.


By the way, you referred to my circle drawing as pure muscle memory. I may have been misleading. The paper, the pencil, and myself are in a visually and tactilely-guided feedback loop. There can be many things in the loop that affect how I draw and how I correct. And no, what I had for breakfast is not relevant, well, except perhaps for the coffee that gave me the jitters.


The point here, Ken, is that human intellectual abilities are far more sophisticated, yet efficient and real-time capable than our model-and-filing-cabinet-lookup analogies suggest. A dead, static entity like a model will not explain anything. As an alternate analogy, I believe some work has been done using neural circuits (simulated in silicon) to (potentially) sort mail by the handwritten zip code. The system had to be trained to recognize handwritten zip codes correctly, but it does not store and “call up” a model of them. The environment has shaped the system into a mail-sorter. Human beings themselves are intelligent. Intelligence is not something that accompanies them and gives them powers.



There is a definition of leprechaun as well. It's an actual model I am trying to chase down here.



You often use the phrase “build a model.” But one must understand airplanes before one could know how to build a model airplane. If one didn't, one wouldn't know what to put into it.



Individual and environment will not get bogged down because the environment, which includes society, encourages or forces the individual to economize. We don't factor in Einstein's breakfast into his theories because they don't make a difference to the outcome.



We aren't discussing science's use of the term “model”, but your use of it. Science's use is primarily for illustrative purposes, whereas you are using it in this thread as an explanatory device. You have models explaining why we do things the way we do. I agreed all along that we can profitably analogize what we do and how we improve to models. You try to reify models, but it is not clear that is necessary for you to say what you need to say.



I know the feeling you speak of here, but in the general, we don't say that one understands just because one feels that one does. Consider taking a final exam. Under your view, the student would already know that he understands and that the purpose of the exam is so that the teacher can infer that the student understands. Generally, however, the demonstration of proficiency is the fact we observe that prompts us to declare that the student understands. Note that understanding doesn't have to exist here in any one place inside or outside the student. Rather it is a word we use to describe particular set of circumstances.