Hi John Are you working from the "Monadic Parser Combinators" paper by Graham Hutton and Erik Meijer? That was written for Gofer a predecessor to Haskell which is similar Haskell but has some slight differences - particularly in Haskell you would really want to make the Parser a newtype then use can make a Monad instance. Here are some of the bits from that paper with the newtype for Parser - unfortunately it adds some clutter. I think there should be a full version converted to Haskell if you do a web search. You will certainly need the sat and char parsers from the paper - I can post them later if you can't find them through a web search: newtype Parser s a = Parser { getParser :: [s] -> [(a,[s])] } result :: a -> Parser s a result v = Parser $ \inp -> [(v,inp)] bind :: Parser s a -> (a -> Parser s b) -> Parser s b p `bind` f = Parser $ \inp -> concat [ (getParser . f) v inp' | (v,inp') <- (getParser p) inp] instance Monad (Parser s) where return a = result a mf >>= k = mf `bind` k pFail :: Parser s a pFail = Parser $ \cs -> [] pSymbol :: Eq s => s -> Parser s s pSymbol a = Parser $ \inp -> case inp of (b:bs) | a == b -> [(b,bs)] _ -> [] (+++) :: Parser s a -> Parser s a -> Parser s a t +++ w = Parser $ \inp -> case (getParser t) inp of [] -> (getParser w) inp [(v,out)] -> [(v,out)]