After giving his famous METHINKS IT IS LIKE A WEASEL computer analogy to evolution in his book The Blind Watchmaker, Richard Dawkins remarks, “Life isn’t like that. Evolution has no long-term goal. There is no long-distant target, no final perfection to serve as a criterion for selection.” I would dispute that there are no targets. Certainly no human has imposed biological targets on nature. But the fact that things can be alive and functional in only certain ways and not in others indicates that nature sets her own targets. The targets of biology, we might say, are “natural kinds” (to borrow a term of use from philosophy).
But let’s grant that the evolutionary process, as governed by the Darwinian selection mechanism, is not goal directed, i.e., that it is not seeking targets (which, of course, leads to the question how a non-directed process is, nonetheless, finding targets in nature). In that case, it makes sense to think of Darwinian mechanism as a grammar-checker — living things must pass the grammar-checker if they get to survive and reproduce.
Now it’s possible to program a grammar-checker to try to generate sentences while at the same time rewarding complexity. One of my colleagues has actually done this (the dismal pattern of anonymity continues since to reveal his identity could undermine his career). As he ran the simulation, he found that the first grammatical sentences he tended to get were swear words (these are one word interjections that pass the grammar-checkers). As he ran the simulation he got more complicated sentences. But not much more complicated and nothing resembling the works of Shakespeare.
No doubt, this negative result is not surprising. But it points up that just as grammars are not sufficient to generate meaningful texts, so the Darwinian mechanism is not sufficient to generate the functional complexity of biology.