Yes, there is operator precedence, but it only matters for infix notation. Or rather, function application has highest precedence, so there is no ambiguity using it.
My broader point was that the space is the most common symbol in English because the function that it performs is the most fundamental action of the language. And since Haskell is a functional language, function application is fundamental, and thus you gain a lot of terseness by using spaces to denote it.
Lisp does the same thing: `(+ a b)`, it just puts parentheses around things to make the parsing rules simpler. If you parse it in a different way, you won't need that many parentheses.
My broader point was that the space is the most common symbol in English because the function that it performs is the most fundamental action of the language. And since Haskell is a functional language, function application is fundamental, and thus you gain a lot of terseness by using spaces to denote it.
Lisp does the same thing: `(+ a b)`, it just puts parentheses around things to make the parsing rules simpler. If you parse it in a different way, you won't need that many parentheses.