Many people claim that scientific explanations of life and cosmology can answer the "how" questions, but not the "why" questions.
That got me thinking about what the "why" question really means. Why actually means how. "Why did the chicken cross the road" means "How did the chicken come to cross the road."
This leads us into a definition of explanation.
When we ask how (or why), we are always asking about the connection between a puzzle and its environment. We are asking what must be true of the local environment such that the puzzle makes sense within it.
The physics questions, of course, fit into this pattern quite well.
"How does this bridge support the weight of the freight train that crosses it?"
There are rules in this environment (e.g., Newton's Laws, and the properties of the elements, steel in particular) that, when taken in combination with the puzzle, form a consistent whole. Given this consistent whole, the bridge can, indeed must, support the weight of the train.
Why questions aren't any different.
"Why did I choose to study physics?"
My parents encouraged my interest in science. When I showed them that I knew about a subject, I was rewarded for my study. In high school, I was good a physics because I had spent so much of my youth thinking about the concepts within it. I also wanted to know the secrets of the universe (who doesn't?). I wanted a starship and a TARDIS, if it was available. I could also get government funding to study the subject. And so on.
All of these environmental factors shaped my personality, and when the time came to decide, I could hardly make any other choice.
By the way, nothing in this analysis prohibits chance events. Consistency is the driving force, not full determinism.
However, there are some questions for which "how" questions make no sense.
When the puzzle in question is the totality of the environment, the questions become meaningless. There is no enclosing environment in which the environment can be said to fit. The environment is just a self-consistent structure, dependent on nothing external.
Here's an example. Suppose we consider a geometric structure governed by certain rules. The structure is formed by line segments in 3D space. Every line segment must meet two or more other line segments at both of its endpoints. There are many such structures, one of which would be the tetrahedron below:
Now, suppose we ask "how is the red line explained by the structure?"
We can say that the environment is such that there are intersections of pairs of lines at both endpoints of the red line, so we have to draw the red line to complete the figure according to the rules of the system. Said another way, without the red line, the environment (the other lines) would not form a consistent system.
Yet, it makes no sense to ask "how is the structure explained by the structure?"
The structure forms a self-consistent whole, and it is meaningless to ask how it is explained by an enclosing structure when there is no such enclosing structure.
Certainly, each individual part of the structure is consistent with its surrounding parts. If we can say this is true of every part of the structure, then we might say it explains itself.
If we observe the universe to be a self-consistent structure with a single, initial event, we need no further explanation of why or how the universe is consistent with anything else.