There are some flavors of consequentialism that allow us to judge something or someone as evil, even if we assume an incompatibilist stance on free will. But that's if there's "harm". If something can not harm, does incompatibilism allows us to judge the thing as evil and immoral? Are there non-consequentialist philosophies that work in an incompatibilist world? Incompatibalism argues that free will and determinism are not compatible. Humans would be like wind or robot, with no ability to "choose" the simplest things like "what to eat for breakfast". It'd all be pre-determined by genetics and environment. Not a "choice". If people do not call wind or robots immoral or evil, because they are not "moral agents", why would they call a human evil or immoral? And if we get to call a human evil or immoral, we should be able to call everything else, be it the wind, river, robot, etc evil or immoral ...
The Wikipedia article on hard determinism also explains the implications for ethics: https://en.m.wikipedia.org/wiki/Hard_determinism
If choice is indeed impossible, then it would be incorrect to hold anyone morally responsible for his or her actions. If this argument holds, hard determinists are restricted to moral nihilism.
Regarding the repeated mention of "robots" and determinism:
Humans would be like wind or robot,
There is not a single kind of robot, just as there is not a single kind of animal. If you think there is a difference between a human and a bacteria in terms of will, then you should stop using the word "robot" in philosophical arguments.
Also determinism is not that important to the question really, even in non-determinism, in a universe where some aliens have true golden free will deluxe with all the extras (the good stuff), it is possible that humans do not have free will.
This is called Epiphenomenalism (https://en.m.wikipedia.org/wiki/Epiphenomenalism), which is the view that or actions are not based on conscious choices, but effects of neutral events in the brain. This is not strictly linked to determinism, even if it sounds similar to incompatibilism. But Epiphenomenalism avoids all flawed comparisons of humans to wind or robots.
Perhaps what it boils down to is the interpretation of ‘free will’; if a human is able to make a choice independently of any external factors then one might argue that this is free will even if the choice they will make would be predictable if one had absolutely precise information about that person. In reality the latter is very likely unknowable and so we can safely go about our lives as though these choices are unpredictable. By corollary we might argue whether a (non self-aware) machine is malfunctioning - at an intuitive level Sharon might say that the oven is broken because it burnt her pumpkin pie, but at a deterministic level we might say that the oven obeyed the laws of physics and so it’s working as it should. The question of whether either opinion is good or bad is a circular argument; if a person can’t be good/bad then anyone’s assessment of them can’t be either.
Whether you're a hard determinist or not, there's still a difference between the wind and people
If I tell you to eat oatmeal for breakfast, you may decline, saying you do not care for oatmeal. Prior conditioning and genetic predispositions and all. If the wind is blowing east, the wind will blow east.
For most people, if I pointed a gun at them and told them they were going to eat oatmeal for breakfast, they would eat the oatmeal. Whether I freely chose to point the gun and make the demand, or whether I was inescapably genetically and environmentally predisposed to point the gun and make the demand, their behavior is still changed by their perception of the threat from the gun! The wind will not blow west just because I threaten it.
We make rules (laws) on the assumption that having a rule (and a punishment) will change the behavior of people. In reality, I won't force people at gunpoint to eat oatmeal, in part because I don't want to get thrown in prison for brandishing a weapon (though I am also fairly indifferent to others' dietary habits).
But whether I freely choose not to threaten people with deadly force to get them to eat oatmeal, or whether I only refrain because of the threat of law or social disapproval... People who do not refrain from forcing others to eat oatmeal, in spite of the law, have to be physically removed from society.
If we could stop tornadoes, we'd stop them. But we wouldn't waste our time with threatening them with consequences for not stopping, because we know tornadoes will not change based on the threat. People's behavior will change based on the threat of punishment
(And if a robot's behavior would changed based on the threat of punishment... We'd probably call the robot a person, too, and say it was immoral for doing harmful things.)
Yes. In an incompatibilist deterministic world where people's actions are decided by their internal rules, then the phrase "free will" can't mean "not determined in advance", which makes no sense, but it can be replaced by the condition "decided by their internal rules".
The human moral system not only implements moral rules restricting our own behaviour, it also contains a mutual enforcement system to examine the behaviour of other people and determine if they're following the same moral rules. People following different rules, or no rules, are called 'evil'.
If an action is forced by external circumstances (they sneezed, causing the gun to go off), then you can't tell anything about the actor's internal moral rules from the act, so it's irrelevant for the purposes of moral policing. This is what we mean by 'responsibility'. If somebody is not responsible for their action (their internal moral rules had no say in it) then you can't judge somebody to be a bad person because an act of theirs caused harm. Moral judgements are a judgement of the correctness of those internal moral rules. 'Free will' means the decision was determined by the content of those moral rules. You can only apply a moral judgement to an act if the person acted in a way determined by their internal moral rules.
Computers in a network can monitor each other's activities, and if one computer fails to follow the protocol, then the other computers might decide it's malfunctioning and isolate it from the network, or shut it down. But there is a difference between a computer not following protocol because it's malfunctioning, and one not following protocol because a cable elsewhere in the network has broken and it simply didn't get the message. Its following the proper rules as best it can, but external circumstances have put it in a situation where it cannot avoid doing harm. Everything here is deterministic, there is no free will in the classic sense, but we can identify a clear distinction between a computer causing harm to the network because of its failure to follow protocol (due to a bug or malware in its own software), and a computer causing harm to the network through no fault of its own.
In a deterministic world, systems of morality arise deterministically. A cooperative network of devices have to jointly follow rules to succeed. Having specific rules for behaviour to enforce the rules on other group members is more fault-tolerant, and succeeds more often. But the rules have to accurately determine if the fault was due to some unit not following the proper rules, or some other cause. Only when it is clear that a unit acted wrongly because it's implementation of the moral rules is faulty is it beneficial to eliminate the unit. We can come to moral judgements without needing a non-deterministic 'free will' - the distinction is replaced by the condition that the actor's moral judgement and compliance was part of the act's cause.
Probably. But it doesn't matter.
The true dilemma of free will is this: Whether we do or do not have free will, we behave as if we do. If we have free will, we behave this way because we have free will, and if we do not have free will, we behave this way because we have no free will.
From the point of view of morals, regardless of whether we do or do not have free will, there are actions that either work towards or work against goals that we would consider moral (such as well-being or whatever). The life history and environment of an individual will have an influence on the actions they take. Therefore, in order for people to take actions that are compatible with a moral goal, they must have a life history and be in an environment where taking those actions is preferable to taking different actions. If people are taking actions that are incompatible with those goals, we have a motivation to stop them.
Therefore, regardless of whether or not there is free will, our moral behavior, motivations, and rules should be essentially the same if our goals are the same.
Does hard determinism leave room for evil and morality? Evil without harm, free will and moral agency?
Yes, but it depends upon your definition of evil.
If by 'evil', you mean intentional acts over which an individual had true agency (ie: the ability to have freely chosen otherwise than the evil act), then the answer is 'no', for hard determinism allows for no notion of non-illusory free will and moral agency requires free will.
But if by evil you mean something like, 'Bad things done by people to other people', then of course the answer is 'yes'. People do awful things to each other and to other animals all the time, and there is nothing about a hard-determinist worldview that inhibits such evil. It merely provides an explanation for it other than true agency.
As for morality, it clearly exists experientially whether or not we have true moral agency. It exists as a concept we routinely engage with; one which has great influence upon our behaviour.
Hard determinism leaves absolutely no room for any kind of moral agency or responsibility.
Hard determinism includes only one chain of causal physical events. Everything mental, moral or psychological is completely excluded.
Any discussion about hard determinism is purely theoretical speculation. Hard determinism is not a thing of reality.