Do I have a choice? (determinism vs libertarianism vs compatibilism)

Yes, I believe it is compatible. But let's concentrate on the "free will" aspect, and skip "understanding", so that we can keep close to the topic of this thread. I expect that single celled organisms are not quite as completely deterministic as you seem to think. But then, I'm not a biologist, so I could certainly be corrected. In any case, we are debating compatibilism, which holds that free will is compatible with that kind of determinism.AI is based on the use of digital logic chips. In principle, an AI system is an ultra-logical system. Logic is supposed to be truth preserving. So if a logic machine receives true inputs, then it will generate true outputs. The received view of truth, is that it is absolute. What is true for you is true for everbody.Here is the problem for the AI system. It makes its "choices" on the basis of logic, hence on the basis of the truth of the result. Since truth is absolute, the decision that the logic system should make is determined by these absolute truth conditions. So all "choice" by the AI system is forced by the truth of the matter. And if the "choice" is dictated by such absolute truth conditions, then the AI system has no real choice at all.Evolution is a pragmatic system. It makes its decisions based on what works, or what is likely to work. Evolved systems (biological organisms) are, in essence, pragmatic machines. They operate by doing what works for them. With pragmatic decision making, we don't have the same sort of absolute standard as people tend to assume for truth. What works for you might be different from what works for me. Moreover, pragmatic choice are not binary. At a certain stage of my life it would have worked for me to become an electrician. It would also have worked for me to become a plumber. There were lots of possibilities. As it happens, I chose to become a mathematician.I think our idea of "free will" comes from our experience, where we are often confronted with an array of choices, all of which could work. Yet we still have to choose between them. Such is the lot of the pragmatic machine.

---

Compassionate conservatism - bringing you a kinder, gentler torture chamber

I am far from an expert on use of genetic algorithms.You can setup rules which specify what will be considered pragmatic. And once you setup rules, you have managed to simulate pragmatic decisions in terms of truth/logic decisions. That's sufficient for simulations of evolution.Some people believe that all pragmatic decision making is really a matter of true/false decisions, based on rules as to what is to be considered pragmatic. Personally, I find that implausible.

What would be the point of free will, if you cannot you it to choose the flavor you prefer?What would be the point of free will, if you could not choose to behave rationally?The usually stated requirement of free will, is that you could have chosen otherwise. You apparently want to instead require that you would have chosen otherwise.Let's put it this way. If we are completely free agents (i.e. have free will), and if we are rational, then we would expect that if we could replay the past we would still make the rational choices, and we would still choose the flavor of icecream that we prefer. If we did not have any free choice at all, but instead a random number generator were wired into our brains to make the selections, then we would behave quite differently on a replay of the past.In short, I am disagreeing with your notion of "free will".

---

Compassionate conservatism - bringing you a kinder, gentler torture chamber

No, it is with the abstractness of computation. It wouldn't matter whether the computation were analog or digital (except that digital is better). Some people say they do, and some people say they don't. I'm in the second group. It is all done in an artificial environment, as part of an abstract computation. In that environment, "works" just means that the programmer says it works.We are drifting away from the "free will" topic of the OP.

---

Compassionate conservatism - bringing you a kinder, gentler torture chamber

I agree that they are not chosen, in the sense of a conscious decision by the child. A newborn child has at most only a minimal consciousness (in my opinion). However, in my view, how the child develops is not completely dictated by the genes either. Learning is a creative activity, not a genetically determined one. I fully agree with that. Responsibility comes from learning to adapt to life in a society. There is no magic instant where one becomes "an authentically free agent". "Freedom" is a relative term. Few, if any, of us is fully autonomous. We are tied to living in a society, and most of us would have great difficulty surviving in the wild, separated from our social support systems. I agree, although exactly what "free will" means is a contentious issue.According to conventional wisdom, at least within philosophy, we are rational agents. I disagree with that. I believe that is a misunderstanding of human nature.I say that we are opportunistic agents. The most central aspect of our cognition is in our perceptual systems. We use our perception to recognize opportunities, and to seize those opportunities as they become available.We are typically faced with a number of opportunities. Often there is no rational basis for selecting between them, although we are very proficient at concocting after-the-fact rationalizations to justify our choices. Looked at this way, I believe our notion of free will comes from this opportunism, and our need to select between opportunities. Our notion of moral responsibility has to do with whether we honor our social obligations as part of our process of selecting between competing opportunities.Incidently my skepticism toward AI, is because AI is attempting to build a rational agent rather than an opportunistic agent. I been known to say, with a deliberate play on the double meanings of the words, that rationality is irrational and logic is illogical.

---

Compassionate conservatism - bringing you a kinder, gentler torture chamber

I can only respond that I believe this impossible. A person is intricately connected to the environment. If you maintain a brain replica in a lab situation, then you sever this connection. The brain-in-a-vat might well die due to sensory deprivation.I strongly disagree with the notion of the person as a stimulus-response machine. If anything, that has it backwards. Life is better described as the person stimulating the environment, so that the environment will respond in ways useful to that person.

---

Compassionate conservatism - bringing you a kinder, gentler torture chamber

I'm sure that a large part of our disagreement is over the meaning of "free will." You have a considerably more expansive view of what it means, than I have.Here, for the record, is my tentative definition:free will is the ability to do science. That is, it is the ability to make the choices necessary for the critical testing of hypotheses and predictions.

---

Compassionate conservatism - bringing you a kinder, gentler torture chamber

We need to distinguish between two notions of determinism. I'll call them "cosmic determinism" and "agent determinism", for want of better names. Cosmic determinism is the idea of determinism in the cosmos as a whole. This appears to be what hard determinists assume, in their argument that what will happen is predetermined so any sense of choice is illusory.I'll use "agent determinism" for the idea that the procedures that control a persons behavior are deterministic, even if the cosmos as a whole is not deterministic.For myself, I don't see a necessary problem in agent determinism. If we want to say that an agent has free will, then we want that agent to be able to determine his choices. So agent determinism would seem to be desirable.On the question of randomness, if we are talking about cosmic randomness, which breaks cosmic determinism, then I don't see that it has the problems that you worry about. However, agent randomness (the alternative to agent determinism) does lead to your concerns. I think a small amount of agent randomness is not really a problem, as long as the agent behavior is predominately due to determined processes.In summary, I don't see that there is any real problem with the kinds of randomness that might exist.

---

Compassionate conservatism - bringing you a kinder, gentler torture chamber

Actually, no. We don't agree.I think we agree that the kind of robot that is made today could not have free will. But that's because we see that the behavior of the robot is determined by its programming. I think I would actually deny that the term "agent" is properly applicable to such robots, so there is no agent there to credit with having made decisions. But what about a person, who does meet the requirements of agency. It seems perverse to credit decisions made in the behavior of that person to any agent other than the person.

---

Compassionate conservatism - bringing you a kinder, gentler torture chamber

I don't see how that answers JavaMan. Your comment in Message 172 is still a mystery to me. What were you trying to say?

---

Just say no to McCain 2008; he abandoned principle when he caved on habeus corpus

But how does that connect back to JavaMan's comment. You seemed to think you were refuting that comment.You are probably making some unstated assumption about cognition and software. But it isn't easy to follow unstated assumptions.

---

Just say no to McCain 2008; he abandoned principle when he caved on habeus corpus