Free will is the philosophical doctrine that our choices are, ultimately, "up to us". Consequently, an unfree action must be somehow "up to" something else. The phrase "up to us" is deliberately vague, and, just like free will itself, admits of a variety of interpretations. We can ask several logically independent questions about free will.
Determinism holds that each state of affairs is necessitated (determined) by all the states of affairs that came before it. In other words, what happens next is completely fixed by what came before. Indeterminism holds that some states of affairs contain elements that were not necessitated by the previous states of affairs. In other words, what happens next is not completely fixed by what came before. The idea of determinism is sometimes illustrated by the story of Laplace's demon, who knows all the facts about the past and present and all the natural laws that govern our world, and uses this knowledge to see the future, down to every detail.
Many philosophers hold that determinism is at odds with free will. After all, if everything that happens is completely determined by the past, how can our choices be free? Wouldn't our choices just be one more outcome determined by the past? According to determinism, we can't just decide to disobey the immutable laws that govern the universe. So if determinism were true, then we would be trapped by the past and free will would be an illusion. This position is usually called "incompatibilism". "Hard determinists", such as d'Holbach, are those incompatibilists who reject free will. "Libertarians", such as van Inwagen, are those incompatiblists who accept free will and deny determinism (this kind of libertarianism should not be confused with the political position of the same name).
Other philosophers hold that determinism is consistent with free will. These "compatibilists", such as Hobbes, often point to clearcut cases of someone's free will being denied -- rape, murder, theft, and so on. The key to these cases is not that the past is determining the future, but that the aggressor is overriding the victim's desires and preferences about his or her own actions. The aggressor is coercing the victim, which is what nullifies free will. Determinism has nothing to do with it. It doesn't matter if our choices are determined by the past, what matters is that our choices are the results of our own desires and preferences, and are not overridden by some external force. This position is typical of compatibilism.
We tend to hold people responsible for their actions. And many believe that one must possess free will in order to be morally responsible. So another important issue is whether we are morally responsible, and in what sense.
Incompatibilists tend to think that determinism is at odds with moral responsibility. After all, how can you hold someone responsible for an action that was bound to happen since the first instant of the universe? Hard determinists say "So much the worse for moral responsibility!" and junk the concept -- Clarence Darrow famously used this argument to defend the murderers Leopold and Loeb -- while libertarians say "So much the worse for determinism!" This issue appears to be the heart of the dispute between hard determinists and compatibilists; hard determinists are forced to accept that we often have "free will" in the compatibilist sense, but they deny that this sense of free will truly matters -- that it can ground moral responsibility. Just because an agent's choices are uncoerced doesn't change the fact that determinism robs the agent of responsibility.
Compatibilists often argue that, on the contrary, determinism is a prerequisite for moral responsibility -- you can't hold someone responsible unless his actions were determined by something (this argument can be traced to Hume). After all, if indeterminism is true, then those events that are not determined are random. How can you blame or praise someone for performing an action that just spontaneously popped into his nervous system? Instead, they argue, you need to show how the action stemmed from the person's desires and preferences -- the person's character -- before you start holding the person morally responsible. Libertarians sometimes reply that undetermined actions aren't random at all, and that they result from a substantive will whose decisions are undetermined. This move is widely considered unsatisfactory, for it just pushes the problem back a step, and further, it involves some very mysterious metaphysics.
Many claim that, in order for an choice to be free in any sense that matters, it must be true that the agent could have done otherwise. They take this principle -- van Inwagen calls it the "principle of alternate possibilities" -- to be a necessary condition for freedom. For instance, if a scientist puts a machine in Bob's brain that makes him kill the President, his action was not free, for Bob couldn't have done otherwise. Incompatibilists often appeal to this principle to show that determinism cannot be reconciled with free will. "If a decision is completely determined by the past," they ask, "how could the agent have decided to do something else?" Compatibilists often reply that what's important is not simply that the agent could have done otherwise, but that the agent could have done otherwise if he or she had wanted to. Moreover, some compatibilists, such as Frankfurt or Dennett, argue that there are clear cases where the agent couldn't have done otherwise, but that the agent's choice was still free: what if Bob really wanted to kill the President and the machine in Bob's brain would only kick in if Bob lost his nerve? If Bob went through with it on his own, surely the act would be free. Or so it is claimed. The problem with this idea is that what Bob "wanted" was determined -- to the extent that it is not random -- before Bob was conceived. Once again, it merely passes the buck to another level. The more complicated and contrived the sense of possibility, the more the principle departs from common-sense notions of freedom.
Throughout the history of science, attempts have been made to answer the question of free will using scientific principles. Early scientific thought often pictured the universe as a highly deterministic place, and some assumed that it was simply a matter of gathering sufficient information to be able to predict future events with perfect accuracy. More recently, developments such as quantum mechanics and chaos theory have introduced greater uncertainty and complexity into the issue.
Like physicists, biologists have also frequently addressed the question of free will. One of the greatest and oldest debates of biology is that of "nature versus nurture". How important are genetics and biology in human behavior compared to culture and environment? Genetic studies have identified many specific genetic factors that affect the personality of the individual, from obvious cases such as Down's syndrome to more subtle effects such as a statistical predisposition towards schizophrenia.
It has also become possible to study the living brain and researchers can now watch the decision-making "machinery" involved in what is commonly referred to as free will. A seminal experiment in this field was conducted by Benjamin Libet in the 1980s, wherein he asked subjects to choose a random moment to flick their wrist while he watched the associated activity in their brains. Libet found that the brain activity leading up to the subject flicking their wrist began approximately one-third of a second before the subject consciously decided to move, suggesting that the decision was actually first being made on a subconscious level and only afterward being translated into a "conscious decision." A related experiment performed later by Dr. Alvaro Pascual-Leone involved asking subjects to choose at random which of their hands to move. He found that by stimulating different hemispheres of the brain using magnetic fields it was possible to strongly influence which hand the subject picked. Normally right-handed people would choose to move their right hand 60% of the time, for example, but when the right hemisphere was stimulated they would instead choose their left hand 80% of the time (recall that the right hemisphere of the brain is responsible for the left side of the body, and the left hemisphere for the right). Despite the external influence on their decision-making, the subjects continued to report that they believed their choice of hand had been made freely.
The theological doctrine of divine foreknowledge is often alleged to be in conflict with human freedom. After all, if God knows exactly what will happen, right down to every choice you make, how can your choices be free? God's already-true or timelessly-true knowledge about your choices seems to constrain your freedom. This problem is related to the Aristotelian problem of the sea-battle: tomorrow there will or will not be a sea-battle. If there will be one, then it was true yesterday that there would be one. Then it would be necessary that the sea battle will occur. If there won't be one, then by similar reasoning, it is necessary that it won't occur. This means that the future, whatever it is, is completely fixed by past truths -- true propositions about the future. And if what will be, will be, why bother trying? Such fatalistic arguments -- arguments that only the actual world is possible -- are usually rife with modal mistakes. A good introduction to the problems of modality is to try to debunk such arguments by isolating their specious entailments.
Some philosophers believe that free will is equivalent to having a soul, and thus that (at least some) animals don't have free will.Determinism vs. indeterminism
Moral responsibility
The could-have-done-otherwise principle
The science of free will
Other issues