Harv, you simply keep missing the issue here. As I have stated several times, I define an explanation to be a method of obtaining expectations from given known information. It is no more than a specific procedure to be applied to a specific collection of information. What is included in the specification "known information" is whatever the explainer chooses to include.

The thing you must understand is that an explanation explains something to someone. Sometimes only one party is involved: that is, you can explain something to yourself. In fact, the second process is the more useful as it is easier to analyze. (The two operations are fundamentally no different). The central issue is communication. The problem of the listener is to understand the explanation. How and when does the listener know when he/she understands the explanation? Why, when his/her expectations are consistent with their experiences. How does the explainer clarify his/her explanation? By adding more information to the communication!

Your concept of an explanation is far too provincial to be completely analyzed. In your picture of the situation, any real explanation of anything involves an assumption that some information is already known. If a model of any and all possible explanations is to be constructed, the model must include all possible information relevant to that explanation including the meanings of the words used to perform the communications.

Total confidence that one understands an explanation is not a possible outcome. If you are to be rational, you must always understand that your understanding may be erroneous no matter how much information you have to go on. The possibility always exists that some added information from further communications will alter your opinion. On the other hand, the listener always has some understanding of the explanation at every stage.

The building of an understanding of an explanation is a dynamic process; in many respects it amounts to a translation of that incoming information. At every stage in the process, one has some personal understanding of the explanation being put forth. My model consists of an exact specified procedure for constructing an algorithm which will yield the best collection of expectations totally consistent with the total sum of known information. That makes it a model of the best understanding of the explanation totally consistent with what is known and totally open to all possible changes.

Since its outcome is a method of obtaining expectations from given known information, it fulfills my definition of an explanation. If you think my definition of an explanation is unacceptable, it seems to me that you need to give an example of an explanation which does not provide a method of obtaining expectations from what is known.

Your only other possibility is to present an explanation which cannot be processed by the procedure I have presented. Now, with regard to this issue, I suspect you are going to once more make massive volumes of assumptions concerning what is known initially. The explanation you present must include all information necessary to establish the proposed understanding of the explanation. No information can be omitted from the explanation.

As I said earlier, the building of an understanding of an explanation is a dynamic process. If the volume of information is small, the algorithm developed yields quite random expectations: an understanding quite analogous to the position, "this is meaningless nonsense". Any explanation you can lay out in a reasonable amount of time will clearly fall into exactly this category. The only reason you feel that statement is not true is that you insist on omitting great volumes of information as already understood.

I hope you manage to understand some aspects of my explanation. I am obviously expecting you to bring to bear what knowledge you already possess together with a little serious thought.

Have fun -- Dick