Some Questions Concerning Free Will
On a post I wrote about free will illusionism recently, there was a latecomer (Jan Luszczek) to the conversation who posed some questions that I would like to look at here, and throw some out for you to answer, and to get the discussion going. English is not his native tongue, so forgive him any inconsistency in language.
I'm late to this discussion, but hopefully someone reads this and cares to respond.
Like many before, since stumbling on the topic of Free Will, I've been trying to reconcile the various writings with my personal "illusion" of choice. As I read up more on the topic a few things kept bothering me. I'll try to do a brief summary here.
1. The linguistic slight of hand as in "you have no control over your actions and what you are". It seems to be a straight forward and clear statement until the arguments for it are put forth. There are three instances of the word "you" in the statement above, and their equivalence is implied, but the arguments don't treat them as such. When diving into the arguments it becomes clear that the first "you" is consciousness, the second "you" is all the functions of the nervous system including subconscious and "house keeping" functionality, the third "you" is the complete physical set of cells and other matter that comprise the body of the subject. Three related, but totally not equivalent meanings of one word. Once individually explained, the arguments never come back to reconcile the first statement.
I care to respond!This is a common problem in philosophy and theory if mind. Indeed, in some sense, one can argue that there is no central "I" or "you". There are arguments, such as I adhere to, that there is no such thing as the "continuous I" - I am a different person than I was as a 2-year-old, a twenty-year-old and who I was even five seconds ago (let alone as a blastocyst!). There is a difference, though, between denying something persisting consistently and continuously over time, to eliminating it entirely. Indeed, no-self theories of mind are eliminativist, so when using general statements involving lots of "you"s, one must be careful, for sure.
This is a common problem in philosophy and theory if mind. Indeed, in some sense, one can argue that there is no central "I" or "you". There are arguments, such as I adhere to, that there is no such thing as the "continuous I" - I am a different person than I was as a 2-year-old, a twenty-year-old and who I was even five seconds ago (let alone as a blastocyst!). There is a difference, though, between denying something persisting consistently and continuously over time, to eliminating it entirely. Indeed, no-self theories of mind are eliminativist, so when using general statements involving lots of "you"s, one must be careful, for sure.
However, for illusionism to hold, or any such position, there must be a perceiving entity to be deceived. We might not be able to exactly nail what the "I" might be, but in a Cartesian sense, something exists in order to perceive itself.
That said, I didn't mention "you have no control over your actions and what you are" in the original article. But the point is salient - whatever the "you" might be at any given time, it cannot have impact and control over the past, and thus that makes the "you" the particular "you" it is (especially given our understanding of genetics and epigenetics). So, careful defining aside, the general point maintains.
2. Dogmatically proclaiming consciousness to be the only possible source of Free Will. In all articles and discussions I've read so far that assumption is at the root of the arguments, even if not specifically so stated at the outset. The problem with it is that even some of the arguments against the existence of Free Will use Libet experiments as proof that we have no Free Will. If a problem is solved subconsciously and the solution is communicated to consciousness, that does not refute the existence of Free Will, it merely shifts its possible nature or origin to functions of the brain that are not directly responsible for consciousness. That would be the second "you" from my first point. Phenomena such as Eureka Moments and our ability to improvise music and poetry show that logical thinking and problem solving can take place without any involvement of consciousness.
I think what is posited here as a conundrum is solved when more carefully defining libertarian free will, as I tend to do when trying to be more refined and diligent: the real, rational and conscious ability to do otherwise in a given situation. If free will is rooted in the nonconscious, and particularly if the conscious self is supervenient on the nonconscious, then there is little to separate such from mere behaviour of a sentient automaton (which , I argue, is in some sense what we are - very complex situation-action machines). Choices aren't being made by a conscious you, but by a layer further beneath, and the choice is only reflected in your conscious mind. In other words, epiphenomenalism (which can be defined as consciousness being the natural by-product of the machinations of the brain such as steam is of boiling a kettle, but it doesn't drive the mind, only merely reflects the physical machinations in a mental manner).
From another point of view, though, since the mental unequivocally depends on the physical (stick a fork in your eye socket and into your brain and see if your consciousness isn't affected), then the origin of free will, if it exists, must be in the physical nonconscious brain.
Phenomena such as Eureka Moments and our ability to improvise music and poetry show that logical thinking and problem solving can take place without any involvement of consciousness.
Absolutely, and this is something I mentioned in my book on free will, but it does not help the free will proponent.
3. Unspoken assumption that thinking and reasoning is a serial and linear macro process much like in a computer processing chip. Give the brain a set of inputs and the output can only take one shape. That is not true even in computer chips. For one, manufacturers do a lot of engineering to ensure repeatability of processing. Two, software can be written is such a way as to not be deterministic. There is a complete subset of information theory dealing with it called Nondeterministic Automata. Billions of neurons working in parallel and interacting with one another resemble nondeterministic automatons much more than x86 processors. The obvious come back from incompatibilists is that randomness is the antithesis of Free Will, but that is true only on the surface. We have already engineered machines and software that deal with noise and error correction in signal transmission and signal processing, why exclude the possibility that our brains have circuits similar in functionality? Obviously we function in non random ways, so the randomness (or rather probabilistic functioning of neurons) is being dealt with somehow. When asked to choose food for breakfast we choose food and not books or cars, but the choice of any particular food item could easily be partially random. The final choice is a function of a system with multiple feedback loops and pattern matching filters. Until we really understand how the brain functions, we can't make sweeping assumptions.
Well, firstly, yes, random is not a good bedfellow for LFW.
The way I understand non-deterministic automata (though I am no expert at all) is that they are not truly deterministic anyway (i.e. completely random). They do the same things as deterministic automata, but in fewer steps. The problem here in computing theory is one of linguistics: "deterministic means non-randomised, but non-deterministic does not mean randomised". And:
In the context of Turing Machines, "non-deterministic" really means "parallel". A randomized algorithm can randomly explore the branches of the computation tree of a non-deterministic Turing machine, but a non-deterministic Turing machine can explore them -all- at the same time, which is what gives it its power. - source.
This also has some useful nuggets.
The question is, since all machines are made of stuff, whether things at a fundamental level are indeterministic. And since we are not yet sure about quantum physics (though the pendulum has swung back in favour of deterministic models from the once preferred Copenhagen Interpretation), this is perhaps unanswerable. A deterministic interpretation of reality certainly makes a lot more sense of causation and probability, I will claim that much.
Either way, I am wholly unconvinced that such indeterminism will allow for the sort of free will that is inferred when talking of LFW. The Information Philosopher, Bob Doyle, makes a big thing of his two-stage model involving quantum indeterminism as allowing for a sort of free will. I remain unconvinced of this as pertaining to the sort of LFW most philosophers mean.
Let's take the commenter's claim: "When asked to choose food for breakfast we choose food and not books or cars, but the choice of any particular food item could easily be partially random."
Free will, to be meaningful in any way (as hinted at earlier) needs to be rational. An irrational or a-rational decision-making mechanism is simply not useful for personal agency and any notion of moral responsibility. To introduce random into the process really hamstrings rational agency. The choosing is, at some level, relying on an internal die roll. The issue then becomes about the wieghting of certain reasons.
For example, if I am choosing to eat toast and not herring for breakfast, I have to weigh up the pros and cons for each choice in a way that is pertinent to me, and not someone else. I am me, and for me, the weighting will favour toast because, perhaps: a) I don't like herring; b) I do not have enough time to make the herring; c) it will make me and the house smell of fish etc. etc. The point being, I weigh those pros and cons up in such a way, and with such weighting (that differs from the next person) because of who I am and the experiences I have had. If I grew up in Iceland living off herring as a result of my father being a fisherman, those weightings would be very different.
Causality in decision-making requires deterministic rationality. As Daniel Dennett has argued, as a conpatibilist, free will needs determinism to be made any sense of.
That should do for now. Here are the other questions and points that you can get stuck into. I will try to return to them later.
4. The lack of clarity in the definition of Free Will. Libertarian Free Will is a logical contradiction and no arguments against it really need to be proposed anymore. But when asked about the possibility that decisions are made in a not fully deterministic fashion the opponents go back to arguing against Libertarian Free Will or repeat arguments that are flawed in ways I outlined in my first 3 points. If Free Will, as spoken about in the article above is an illusion we should be able to empirically test for it. We should be able to come up with an experiment that shows the feeling to be false. But for that we would need to first define how a theoretical being possessing at least some degree of Free Will would be able to act. I have not come across such proposal yet. I think that it's because a theoretical person possessing at least some degree of Free Will would be empirically indistinguishable from an average human.
5. This is somehow related to point 3 but deserving its own little paragraph. Without straying into pseudo scientific, New Age mumbo jumbo, there are serious scientific theories that include quantum effects in the functionality of living organisms. Apparently, birds "see" magnetic fields, the sense of smell could be sensitive to quantum spin because two otherwise identical molecules can smell differently and the only difference between them is the spin of some electrons. (EDIT: Also photosynthesis apparently owes its efficiency to quantum effects.) If neurons that take part in thinking possess some quantum receptors, determinism of choices would go out the window and we would have to reconcile what we know about consciousness and thinking with the new facts. I bring this up only to signal to the participants in the Free Will debate that there is a lot we need to learn about the nature of thinking and consciousness before making sweeping statements about them. In my opinion neuroscience might give us the true answers, but that is many years away.
6. The arguments for and against Free Will might never be answered conclusively. The work of billions of neurons and trillions of synapses might just be beyond the human ability to comprehend, just like chimps could never understand quantum physics. Maybe AI will one day become sophisticated enough to grasp it and give us the essence of the answer.
Over the years I thought of some other flaws in the debate and maybe if challenged on the ones above I will recall some of them, but for now these are all I can think of.
[caption id="attachment_6955" align="aligncenter" width="550"] My first ever book. Please grab a copy![/caption]