The understanding between a non-technical writer and his reader is that he shall talk more or less like a human being and not like an Act of Parliament. I take it that the aim of such books must be to convey exact thought in inexact language... he can never succeed without the co-operation of the reader.
It is also a good rule not to put overmuch confidence in the observational results that are put forward until they are confirmed by theory.
Science is one thing, wisdom is another. Science is an edged tool, with which men play like children, and cut their own fingers. If you look at the results which science has brought in its train, you will find them to consist almost wholly in elements of mischief. See how much belongs to the word "Explosion" alone, of which the ancients knew nothing.
[When thinking about the new relativity and quantum theories] I have felt a homesickness for the paths of physical science where there are ore or less discernible handrails to keep us from the worst morasses of foolishness.
I don't believe any experiment until it is confirmed by theory. I find this is a witty inversion of "conventional" wisdom.