What does understanding mean ?

I'm attempting to make a definition of understanding. Because I think this is as primary as it is useful. We need understanding when we feel baffled, i.e., we feel that our pure logical deductions can not make an immediate connection between what we know (as a part of our experience) and what we just observed (which is the object to be understood). In other words, we feel that, these two ends, our knowledge as one end and the object as the other, seem very distant. Our logical deductions seem helpless in ferrying us from one end to the other. As long as this ferry is not finished satisfactorily, the understanding will be on-going. The process of understanding is to build these logical steps (the causal chain), starting either from existing models or a new one, all the way to reach the object. So, only when (1) the proper model has been found and (2) the causal chain has been forged can the understanding be settled, and can we be released from the baffling. Our baffling, in my opinion, is a gift. The sense of being baffled promotes us to raise questions, and raising questions reverts us into another baffling. We are baffled when we have questions. As long as we have questions, we shall be baffled. This is a joyful voyage. That is why Einstein once remarked, 'the best one can have in life is the experience of mythtery, and I'm content with a life of mythteries'.

An example. If a man fucks a woman, this woman may get pregnant. So, the question is, 'why does she have to get pregnant ?' 'Why cant it be otherwise ?' . 'Why cant it be otherwise ?' is a question that urges one to answer. We know a fucked woman may get pregnant, but we don't know why this simple fucking could lead to a baby ! If one asks himself, he'll be baffled and curious. It is never self-evident that a fucking shall bring about a baby. An understanding may be achieved if he finds, 'fucking ——> ejaculation of semen ——> semen entering the womb ——> semen's synergy with an egg in the womb ——> this compounded object divides and grows ——> a baby forms in the womb'. If this chain is found, his baffling shall be more or less alleviated. However, only after he completely confirms every element of this chain will be fully relaxed.

In the same spirit, I wish to talk about computer simulations, which have become a very important tool in theoretical physics and other fields. It often offers very important insight that may lead to ultimate understanding, albeit it is not an understanding by itself. It helps understanding, just as experiments. Actually, computer simulations play the same role of experiments, I reckon. In experiments, you set experimental knobs and then start experiments and observes what will happen and make record. In computer simulations, you set and input required parameters and then let a computer to execute orders and output results and you record the results. The only difference is that, in the former it is the Nature that composes and executes the orders while in the latter it is you who write the codes to be executed by the computer. After simulations or experiments, you get the outputs. But you don't know why the output looks like this but not like that. The causal chain between the input and the output is not clear and awaits building. Frequently, this chain can seldom be exactly built. Many approximations have to be made. Much the way one builds a bridge. For the bridge to be strong, perfect materials should be used. But perfect materials can hardly be found, so instead one uses the best at hand. 'The best' may not be perfect, but at least a bridge can be laid down. When better materials are found, an improved one can be built.

That is the way science is done, I think.