Now back to your original problem. Can you write a function g such that g [1,2,3,2,4,1,6,3,6,2,3,5,2,5,2,1,6,4] returns ([1,2,3],[2,4,1,6,3,6,2,3,5,2,5,2,1,6,4])
g [2,4,1,6,3,6,2,3,5,2,5,2,1,6,4] returns: ([2,4,1,6,3],[6,2,3,5,2,5,2,1,6,4])
and so on ? Then you should be able to build what you wanted by using g in a modified version of f.
For this kind of questions, Hoogle[1] is your friend. You have a list of type [a], and you want to split it into a pair of (the list up to some point, the rest of the list), whose type is then ([a],[a]). So you search in Hoogle: [a] -> ([a],[a]) and the following results come up: Prelude.break :: (a -> Bool) -> [a] -> ([a], [a]) Prelude.span :: (a -> Bool) -> [a] -> ([a], [a]) (... a few more...) A quick review of the descriptions will show that one of these functions is /very/ close to what you want to achieve. You may use it to implement your solution and/or look at its source code to understand how it works. In fact, you can do better---look at your /original/ problem and figure out what type to search for in Hoogle. One of the results there should turn your problem into a one-liner. Again, you can take a look at its code for additional insight. References: [1] http://www.haskell.org/hoogle/ Have fun! -- Ariel J. Birnbaum