OK, so you don't believe in "evil", but you believe in "barbaric".
I believe that determinism and free will are entirely compatible. An algorithm has free will. More complicated algorithms have a more nuanced, richer free will. That's it.
For instance, a coin-operated machine that gives you a bag of potato chips in exchange for a dollar has a form of free will. It does that because it wants to. It is just not capable of telling itself it wants to do anything else; it's a low grade form of free will encompassing a tiny number of states.
(Killing evil is basically just terminating a buggy algorithm. If you don't like the baggage associated with "evil", maybe "defective" or "buggy" is better.)
The question of free will and determinism is made complicated by the possibility that a deterministic free will (algorithm) operates in a world that isn't deterministic. Suppose that your mind is an algorithm which will make exactly the same decision for the same inputs (including, of course, its own state: that's one of the inputs). Even if you get your mind to be in exactly the same state as when a certain decision had been made, you also need the world to be in exactly the same state. Only then is the algorithmic mind guaranteed to think the same thing and make the decision.
The world has so many states, and so does the mind, that the question of whether you are algorithmic or not makes practically no difference. Even if you are a FSM, your state space is so large, you will never be in the same state twice -- just due to the fact alone that you have life long memories that are are still accumulating, for one thing. And if you could rewind exactly a previous state, the surrounding world will not; and that has an even vaster state space.