Actually, Haskell lets you do it either way. It's the best of both worlds :)

This code:

    do a <- [1,2,3]
       b <- [4,5,6]
       return $ a + b

can also be written as:

     do {
       a <- [1,2,3];
       b <- [4,5,6];
       return $ a + b
     }

or:

     do { a <- [1,2,3]; b <- [4,5,6]; return $ a + b; }

I've even seen it like this:

     do { a <- [1,2,3]
        ; b <- [4,5,6]
        ; return $ a + b }