If the hard determinist position is true, then no-one can really be held responsible for their actions
Of course they can. That's subjective.
On the other hand, if the libertarian position is correct, then the deterministic model that underlies science must be wrong.
My intuition, like that of most people, is that this argument is faulty in some way. I spend a good part of my day putting a lot of effort into choosing between alternatives - it doesn't make much sense to imagine that these choices would have been the same without me putting in the effort to make the choices.
You're right -- it doesn't make much sense to imagine that, since such assumes that you can change the past.
Now that I've got to the end of this essay, am I entirely free to post it or not to post it?
No. "Entirely free" is self-contradictory; as, "that which is locked to nothing," wouldn't be locked to being 'entirely free'. Oh, and it wouldn't be locked to that logic, either. Nor that logic. Nor that.
Ain't violations of the law of noncontradiction grand?
Or is my action of posting it (or not posting it) already predetermined as I write these words?
Is there some entry of true randomness? If not, then yes.
(a) You often hear hard determinists talking about man being an automaton. However, an automaton is a thing that follows a pre-programmed set of instructions, and any organism with a brain works in quite a different way.
But the way in which we work only makes us more so. We learn automatically -- our programming changes on-the-fly. For the things normally referred to as 'automatons', this is not automatic.
(b) The hard determinist model fails to distinguish between organisms that can make choices and those that can't. The brain is an organ designed to make choices, to separate the organism from immediate stimulus-response mechanisms.
If it is nothing but a conglomeration of immediate stimulus-response mechanisms, why make a distinction? The only difference between a single-step and multiple-step is that the increased complexity of the multiple-step can result in you losing track of exactly what's going on in the middle, so you'd fill the middle in with a question mark -- which gives the concept wiggle room. But your concept of the system including wiggle room doesn't mean the real system has any.
The possibility of a slight change in random components, that could have affected the outcome, does not in any way relieve the driver of responsibility.
The problem is that your random component isn't anywhere in the driver's mental system (which consists of inputs and configuration.) Change it so that it is, and watch what happens:
One of the drunk driver's inputs is changed. Instead of the driver being sent an input consistent with that of light reflecting off a child, the universe randomly outputs the image of a school bus in his path. He swerves to miss it (something that wouldn't have happened if the universe wasn't fucking around with his inputs), and he hits the child.
If the universe sent him bad data, is he responsible for the result?
I'm not convinced the expression "nonrandom set" has any meaning.
It's the expression of nonrandomness.
The orbit of the moon is usually considered to be nonrandom, but it follows no formula.
To model the moon's orbit at 100% complexity, you'd have to take into account every piece of matter's effect on every other piece of matter within a sphere 27.4 billion ly across. The fact that we can't do this doesn't mean the universe can't.
Effects are locked to their causes. Thus, the state of future things is locked to the state of past things. Everything in the past is in a set state; therefore, everything in the future is in a set state. Set = locked into immobility. Anything locked into immobility isn't free.