Any conclusions regarding the nature of any free will we might have are obviously contingent on the nature of the universe, so before even touching the murky topic that is free will I think it first important to figure out what the hell kind of world we all live in. And yes, I do think we all live in the same world. I see no progress being possible in that mumbo-jumbo pomo world of fluff, just as I see there being no hope of persuading any person who believes such nonsense how wrong they are. Instead, I think it wisest to adopt the empirical, scientific viewpoint. Perhaps there are such things as souls, or some non-material part of the mind, but until someone comes up with a clear explanatory ontology for them, or at least some purpose or function they fulfil, I see no reason in postulating some arbitrary ad hoc explanatory tar pit.
Thus we're taking our world to be a purely material world, and one in which science provides good explanations. On these grounds, consider initially the “seat” of the mind, the brain and parts of the spine. One of the primary kinds of elements in the mind are therefore neural cells. As well as this, there are of course various blood-pumping cells and so forth, and there are several types of neural cell. Speaking in general about neural cells though, we know that one weighs, on average, about 1×10-8kg, or 0.000001g1). From this, Avogadro's Constant (6.02214199×1023mol-1) and an estimate of the atomic weight of an atom in the brain cell, we can determine how many atoms there are in a single neuron. Lets assume the average atomic weight of an atom in a neuron is 25. Remembering that the human body is primarily hydrocarbons or water (and thus primarily oxygen, carbon, or hydrogen) will show how absurdly conservative this estimate is. These three atoms have the atomic weights of 16, 12 and 1, respectively and approximately2). Given this, and the fact that they make up the vast majority of the body, for the average atomic weight of any group of neurons to approach 25 would require that they be the sole consumers of the various heavy metals and exotic molecules our body needs, such as lead and arsenic. And as I'm sure you're aware, when we get too much of these they lead either to madness (particularly telling in this context) or death. Having hopefully established these grounds as overly conservative, we can thus trivially calculate that the average neuron contains at least 24,088,567,960,000,000, or roughly 2.4×1016, atoms. Given this magnitude, it should be fairly clear that our brains operate in the Einsteinian universe, sufficiently insulated by their vastness to be relatively immune to the vagaries of quantum level effects. This thus translates trivially to determinism. In other words, from the state of the world at this moment in time, or in fact at any moment in time, it is theoretically possible to predict precisely the state of the world at any future time. Note here though the very specific usage of the words “theoretically possible” in philosophy. It does not mean that we will ever actually be able to achieve this - it merely means that, hypothetically speaking (so to speak), it should be possible to build a computer or machine which could predict the state of the universe at any time from any given previous state. Fairly obviously, such a computer is not possible - it would presumably need to contain more matter than the universe itself, and getting the relevant data to it is simply not possible either. This is largely but in part because measuring the position (or velocity) of the various sub-atomic particles which constitute our universe leads to obtaining the other measurements being impossible, due to their quantum nature. The averaging out of the various unknowable quantum variables that occurs at the macro scale of the neuron and which make it statistically and unavoidably deterministic clearly does not occur at the level of the individual particle. The universe is thus deterministic, but not determinable.
Free will, whatever it is that we mean by that, would thus seem to be in dire straits. We're all just fanstastically complex machines, any future actions unchangeable and unavoidable. This is a topic I will return to momentarily, but I think it first important to look at what happens if my above analysis is wrong. What if the universe is instead indeterminate? Although this might initially appear to be a far better environment, one in which free will can prosper, a closer analysis will show that sadly this is not the case. We are lucky then, that the universe is deterministic. Consider what it would mean if quantum effects were somehow magnified up to the macro scale of the neuron, or, for the purposes of this thought experiment, to the level of something easily human detectable, like a thrown brick. The precise level at which we magnify this indeterminacy too in this thought experiment is irrelevant, because no matter the level it is magnified to, its impact will eventually scale up to the level of the thought experiment I will now go over. Return to the world mentioned earlier, indeterminate even to the level of the thrown brick. Imagine that you are standing in your home (a home presumably under construction) with your spouse when a robber comes in and demands that you give them all your money, and some of your bricks too. Enraged by this audacity (how dare they go after your bricks!) you hurl one of them at him. Unfortunately though, indeterminacy takes hold, and the brick you flung, and which would have (in a deterministic world) killed him, flies sideways and hits and kills your spouse instead. Was this your intention? Was this your decision? Are you morally responsible for killing your spouse? It's not an accident in the normal sense of the word as, at least in our world, the causal chain involved in an accident can be traced back and causal blame (and in some cases, moral responsibility) can still be assigned. My intuitions are that, no, you could not be responsible for something which was entirely not your fault. After all, if the quantum dice had rolled differently (which they could well have done) you would be a hero for stopping an evil brick-stealing criminal. An indeterminate world may have free will, but (whether it exists or not there) it can only be meaningless. Thus my conclusion that free will in a deterministic world is always preferable, even if more limited in its meaning that it might be in an indeterminist world.
But what free will can we have in our deterministic world? Choice seems to be the core element of what free will really is, and it seems to be both necessary and sufficient to entail moral responsibility. After all, if you could not have chosen differently, how could you be morally responsible for what happened? A thought experiment in favour of the position that you could be though runs something like as follows, however I and others find it less than convincing: John does not know it, but there is an evil magic demon who watches over his shoulder throughout all his life. This EMD has an evil (I bet that was surprising) plan for John, and no matter what John actually choose to do, he will always actually choose and think he has chosen himself to do whatever is nastiest and most hurtful to the others in that situation. Fortuitously though (for the EMD at least!), it never has to use its power, as John always chooses to do the nastiest most hurtful thing to others anyway. Even though John had no choice, to my intuitions at least, it initially seemed that he is still morally responsible for the evil he has done. Conversely, Joe, who always does the best and most helpful thing without her Angel having to force her to, seems just as morally upright as if she had had the choice as to what she did. Clearly, choice is not needed for moral responsibility, no? However, why I no longer find this argument convincing is due to the fact that I think it equivocates over the meaning of the word choice. See, the thought experiment still assumes that John and Joe actually had the ability to have chosen differently - they merely didn't. Thus even though they never could have done differently, they could still have chosen to do differently - even if the demon or angel would then have changed their choice and still made them think the new one was theirs. In a deterministic world, the very concept of choice loses this meaning. Neither of them could ever have chosen differently anyway, and thus the demon or angel was unnecessary. Hence the argument fails to convince, and determinism still seems to rule out, not just choice, but moral responsibility too. Before moving on though, I think it necessary at this point to clarify what we mean when we say: “I could have chosen differently.” Assuming for now that this statement is true, (although we know that it is not) what it should mean is something like: “if I were to find myself in exactly the same situation again, with all other intentional agents involved making exactly the same decisions, and with all things, both intentional and non-intentional, having precisely the same motion and location vectors, and with me having exactly the same knowledge and personality and so forth as I had when I made that decision the first time, I would still be able to make a different decision. If I have the power of choice, I would not be so constrained that I could only make the same decision again.” While this more accurate phrasing of what we mean by choice will lead us shortly to a view on free will which gives us all that we actually want, I think it important to highlight that this longer version is of course not what most people actually mean when they say “if only I could go back…” What they mean is (understandably) “if only I could go back, knowing what I know now…” This is a reasonable thing to wish for in many a situation, but it is a dramatically different thing than the meaning of choice actually entails.
So what meaning of free will is there then? It would seem at first that there is no possible definition of free will that could resurrect in us the hope and life that a belief in choice brings with it. I can honestly say that for quite some time I empathised very strongly with Hume and his recourse to drink, the company of friends and backgammon when he first conceived of Humean bundles. That said though, I must admit that I found backgammon no more appealing than I usually did… Yet after some reflection and though, I came to realise that a perspective just as appealing and far more truthful than a belief that is the myth free will should be was available. Return now to the more complete version of what we mean when we say “I could have chosen differently” and then ask yourself the key question: why would you make a different decision the second time? Certainly, you might if you knew the consequences of your decision - but of course, by definition, you do not. What possible motivation could there be for making a different choice? I can think of only one, and that is simply that you might not have made the best possible decision you could have at that time, given your psychological and informational state. Perhaps you made a bad decision when you were drunk - but it would not have been the same situation if you were not drunk. Perhaps if you had had more time to think, you might have made a better decision. But again, it would not be the same situation if you had had more time to think. I can, in my own case, think of no situations where I have deliberately and consciously selected a decision which would come with what I expected would be worse consequences than another - and even if I had, realising what I realise now about the universe, I would merely resolve myself to only make the best decisions I could at any time. Perhaps I might still bow to peer pressure, or the effects of alcohol, or greed, or any of a number of influences. I claim not that I can or would make the best decision - only that I would make the best decision I could make. This radical reconception of what a human being is - merely a constraints satisfaction machine, striving to satisfy those constraints it finds itself facing as best it can - leads to some changes to moral responsibility. After all, those who are guilty of moral wrong could never have done differently. But that does not mean that they should not have done differently. They had no choice, but it does not change the fact that the constraint satisfaction machine that they are simply does not work as well as it those of others. Time in a correctional facility, or the response and disgust in others is still, phenomenally, a punishment. More importantly, it will hopefully educate them and cause the decision making machine to work better in the future, and at the least (with time in a correctional facility or due to the things others now know about them), it removes such machines from an environment where they can cause harm for some time.
Choice is gone, but we never needed it anyway, and moral responsibility survives, perhaps slightly modified, but just as central and as potent in all of the old places as it used to be.