In his book Grammar for Grown-ups, Craig Shrives attributes the under-mentioned meaning to the following sentence:
Manx cats which live on the Isle of Man have a longer life expectancy than normal domestic cats.
Craig gives this sentence the following meaning: ‘means that only Manx cats living on the Isle of Man have a longer life expectancy' … (‘i.e. Manx cats that live elsewhere do not’ …).
If his interpretation were correct, it would lead to the absurd meaning that the following sentence demonstrates.
Boys who go to King Edward VII School do well in life.
According to Craig's book this would mean: Only boys who go to King Edward VII school do well in life; boys who go to any other school do not.
As I am battling with some of these clauses, I should be grateful for any assistance.
I think you might be extrapolating this ruling like language were a science. It is, of course, an art, and the context needs to be considered.
"Boys who go to King Edward VII School do well in life" is fine. I do not think it carries the suggestion that those who don't go to the school don't do well.
"Manx cats which live on the Isle of Man have a longer life expectancy than normal domestic cats" does, however, carry the suggestion that Manx cats living off the island do not enjoy the same longevity.
I can attribute this difference only to context. I think the sentence below, which is really similar to yours, carries some of the suggestion you're concerned about, but, in my opinion, yours doesn't:
"Deaf boys who go to King Edward VII School do well in life."
It's a great question. Sorry to answer it with a non-scientific "it depends on the context", but, as I said, language is an art.
I agree with Craig, and would just add something here about terminology.
Your question highlights the issue of why the terms 'restrictive' (the type of clauses with no commas) and 'non-restrictive' (the type set off with commas), as used in traditional grammar, are misleading, and that 'integrated' and 'supplementary' respectively are better:
(1) Manx cats, which live on the Isle of Man, have a longer life expectancy than normal domestic cats.
(2) Manx cats which live on the Isle of Man have a longer life expectancy than normal domestic cats.
Here the writer had two choices: either, as in (1), insert commas after "cats" and "Man", thus making it a supplementary (non-restrictive) relative clause, OR omit them entirely making it integrated (restrictive), as in (2). And that choice would dictate the meaning of the sentence. They consciously chose option (2) without commas, and you have to assume this was purposely done in order to give the restrictive meaning of "only Manx cats which live on the Isle of Man (and nowhere else) live longer than normal domestic cats". There's nothing wrong with that, and nothing misleading here about the use of the terms 'restrictive' and 'non-restrictive'.
BUT now for the bit you're really interested in: your school example. It's different:
(3) Boys who go to King Edward VII School do well in life. (integrated /restrictive)
(4) Boys, who go to King Edward VII School, do well in life. (supplementary/non-restrictive)
Again, the writer had two choices, integrated (restrictive) as in (3), or supplementary (non-restrictive) as in (4). They chose (3), which we can show to be the correct choice for two reasons:
(i) If you include the commas (4), thus making it a non-restrictive relative, the implication is that 'boys in general go to King Edward V11 School', and that 'boys in general do well in life'. Clearly, both those implications are nonsense.
(ii) Although omitting the commas in (3) make the relative clause integrated, it's not actually restrictive because there is no implication that only boys who go to King Edward V11 School do well in life. The integrated construction simply provides a succinct way of saying that boys who go to that particular school do well in life, which is no doubt true.
And that is precisely why the terms ‘restrictive’ and ‘non-restrictive’ can be misleading: the integrated relative clause is not always restrictive, as (3) clearly demonstrates.
