>>67470
>The idea of free will is an incoherent idea because it proposes that no mechanism underlies one's decision-making.
This is functioning under a very popular yet nonsensical conception of what free will is. Free will, properly understood, is merely the ability to make decisions. It is the capacity to select between alternatives. If we define it such that it denies any influence whatsoever on that decision, then we have trivially defined it out of existence, but that is merely wordplay. You rightly recognize such a conception of free will as being utterly without any logical merit. I put it to you that anyone advocating for it under such a definition is not engaged in anything that could properly be considered philosophy.
Having addressed the meaning of "free will", let us address what it means for something to be "deterministic". There are two senses in which this word is typically used, and when those senses are used interchangeably, you get the sort of confusion we see in this discussion.
In the first sense, a process is "deterministic" if its outcomes are wholly causally determined by natural–that is physical–conditions. In this sense, all events which occur are deterministic. No effect proceeds without cause.
The second sense, which is much less precise and probably shouldn't be used much at all, holds that an event is "deterministic" if we (or any observer) is capable in principle of predicting its outcome given sufficient data about its initial conditions. In this sense, no process is deterministic at all scales, as some degree of uncertainty must necessarily exist in all initial conditions. Chaos theory shows us that many processes are so sensitive to initial conditions that we cannot acquire enough data to accurately predict the outcomes beyond some rather wide tolerances. We refer to these events as "random". Some processes are even infinitely sensitive to initial conditions, meaning that any difference, no matter how infinitesmally slight, will result in a significant change in outcomes. Such events are still "deterministic" in the sense that their outcomes are wholly the product of physical factors, but they are not "deterministic" in the sense that we can predict them, because we fundamentally cannot.
>>67473
>They only assented to abstaining from revenge because the state promises to enact revenge upon the criminal.
You don't think there's a general causal process which tends to favor pro-social behavior in the absence of state mechanisms? Don't you think humans have evolved a general tendency to try to stay on everyone's good side, possibly because despite your drive for revenge, you recognize that other people have similar drives that could make your seeking of it worse for you in the long run?
>>67475
>This is your brain on atheism.
For the record, I'm an atheist but not a postmodernist, nihilist, polylogist, or anything of the sort. The absolute nature of truth and logic informs my rejection of divinity and comprehension of ethics and free will. There is a lot of philosophical variance among atheists. We're not all against you.
>>67477
>because human behavior is predictable
At what scale, though? Yes, you can predict trends, but you can't really predict what a given human will do anywhere near as well as you can predict what a large group of them will do. When you view chaotic systems in sufficiently large quantities, you will see statistical distributions, but that merely quantifies the degree of unpredictability of a given event. There must necessarily be some finite distribution because there is only a finite range of possible behaviors. That statistical "predictability" is merely a mathematical representation of the ground conceded to unpredictability, and trivially must exist on some scale.
You simply cannot predict the outcome of a fairly-rolled die, no matter how much information you have, because it is infinitely sensitive to initial conditions. You can, however, that a million such die rolls will tend to float around an average of 3.5, because of the range of possible outcomes. You can only get a number between 1 and 6, so trivially the statistical distribution must lie somewhere between those values.
Humans are much more complex than that, with a much wider range of possible behaviors and a much greater dimensionality to those ranges as well.
Of course, even if you could predict a given person's actions with perfect accuracy, that wouldn't mean their will isn't free; it just means that you have perfect information about their decision-making process. Knowing what someone is going to choose doesn't mean that the choice does not occur; that's an unreasonable criterion for defining "choice".