This kind of cheerless wandering around web led me to read some postings and articles about free will. Even some physicists have now accepted "free will" into their vocabulary. The fQXI conference about the nature of time held in some boat sailing from Norway towards Copenhagen las summer had inspired several blog postings. Also I wrote comments about the excellent lecture of David Eagleman about perceived time. This kind of sailing trips cost and it is good if they induce interaction between people with different backgrounds: now at least physicists, neuro-scientists, computer scientists and philosophers were solving both the problem of time and the problems caused by sea sickness at one blow.
I did not find it surprising that I did not find anything surprising in these postings. The common feature of all these articles is that quite too much is assumed. Sean Carroll as a descent reductionist makes especially strong assumptions. All writers have managed to remain unaware of the dramatic distinctions between subjective time and the geometric time of physicist. They also make the same error: in the process of trying to understand free will scientifically their first step is to carefully eliminate conscious mind from the picture. The outcome is free will as something effective and emergent or free will as resulting from deterministic but non-predictable/non-computable process. My humble question is: Why on earth something very complex or non-computable would generate sensation "I decide to do this!"?! A non-deterministic behavior serves as a correlate of free will but non-predictable (but possibly deterministic) behavior does not imply experience of free will.
Every writer grasps something essential but fails to see the essence of the problem and connect it with many related problems like the puzzle of time and the basic paradox of quantum measurement theory. One should not be however too critical since the position of the writers is unrewarding. Being names in blogo-sphere they have been invited to solve the problem of time with minimal background: this is like solving some deep problem in topology with the background given by a couple of hastily read Wikipedia articles.
I was a little bit disappointed but understood that I must also realize that the understanding free will is as difficult as the understanding of the nature of time. It requires a lot of time and a flash of genius: a sea trip from Norway to Copenhagen with National Geographic Explorer - even in a good company - need not be enough to spark this kind of flash. I have been trying to communicate more than 15 years my own flash of genius relating to free will and the relationship between experienced time and the geometric time of physicist but it seems that this has been waste of time. They must discover it themselves! Let us hope better luck during the next cruise! In the following some more detailed comments about articles of the peopled who participated the trip.
Sabine Hossenfelder: Free Will function
Sabine Hossenfelder has a posting titled "Free Will function". I agree with Sabine that the idea about emergent free will is self deception. Free will does not emerge from a deterministic microscopic dynamics.
The believers in emergence say that free will is an effective concept. Not real but useful. If the system in question is complex enough and behaves non-predictably as seen by outsider one can say that it has effective free will. But why the impossibility to predict a deterministic dynamics in practice would generate the experience "I will do this!". There is absolutely no justification for this belief.
A good objection against this identification comes from neuro-science and is described in the article The Brain on Trial by David Eagleman. People suffering Tourette's syndrome, split brain patients, persons with side personalities, and patients with choreic motions behave from the point of view of outsider as they would have free will. Using biblical language: they act as if being possessed. They do not experience free will. Who wills? Who uses the biological body of the patient? Same questions can be made in the situation when people who have done mass murder become conscious and begin to wonder what these bloody hands have done. Who used these hands? Are we merely our brains and bodies? Who uses my biological body? What is this "me"? Is this particular biological body used only by single intentional agent, by single "me" only? I could continue by telling about the notion of magnetic body but will not do it here.
Acidic out-of-topic side remark: Effective theories have become the basic curse of theoretical physics today. No-one can seriously claim that string models say anything about the world of experimental physicists. But there is a loop hole. By postulating effective field theory approach one can build entire landscape of effective theories. This is non-sense but it works. The only honest reaction would be to admit that string models are nice theories but not theories about the world we live in.
Sabine Hossenfelder suggests as a solution something that she calls free will function. Sabine considers a machine spitting out digits of π. This process is fully deterministic but outsider has no means of predicting what the next digit will be and what number the digit sequence represents unless he manages to get the program code. The proposal is that our brain has this kind of free will function. The strange assumption is that the inability to predict would in some mysterious manner generate experience of free will. But Sabine a physicist has learned that one must forget all subjective elements when doing science. In this mental framework the only conceivable goal of a theory of consciousness is to eliminate it. The fruitless searches of "consciousness modules" assumed to reside somewhere in the brain are fruits of similar "consciousness as function" thinking.
Sean Carroll: Free will as real as baseball
Also Sean Carroll has written about free will in his posting "Free will as real as baseball". Sean belongs to the effective theory camp and sees free will as a convenient tool of description just like baseball is seen by a reductionist as a convenient abstraction to describe the dynamics of a condensed matter system.
Sean makes two very strange claims.
- The first strange claim is that free will is inconsistent with the laws of physics. This is the case only if the experienced time and geometric time of physicists are identified. The are not one and the same thing as even child realizes. Experienced time is irreversible and there is no subjective future. Geometric time is reversible and future and past are in the same position. In general relativity 4-D space-time region becomes the basic entity instead of time=constant snapshot which is the basic entity according to Newtonian thinking. Amusingly, all writers except Scott Aaronson seem to belong to the species of Newtonians as far as their views about time are considered. The first years of scientific education involves really heavy social conditioning and it is really hard to de-learn even the obviously wrong beliefs.
- The second strange claim of Sean Carroll is that the physics is completely understood in everyday realm! Really! Do we really understand the physics underlying living matter?! I cannot do help it: this kind of loose text book statements irritate me that suddenly the dull depressive mood has gone and I am full of adrenaline!
Interestingly, Sean Carroll notices analogy of poorly understood notion of free will with the poorly understood notion of time. The arrow of time is in conflict with microscopic reversibility but - according to Sean Carroll - physicists do not see this as a problem so that it is not a problem. Continuing in the same spirit: if billions of Chinese believe in communism then marxism is the only correct view about society and is indeed law of Nature! The effective theory solution is simple: also the arrow of time somehow emerges. Exactly how?: this we do not understand but it does not matter.
This is self deception. One should admit this and really try to understand second law. If one does this, the first observation is that Boltzmann's equations are deduced by assuming the occurrence of state function reductions in time scale much shorter than the time scale of observations. State function reduction is what makes quantum physics non-deterministic at the level of single quantum system - also internally in-consistent: the determinism of Schrödinger equation is in blatant conflict with state function reduction if one identifies experienced time with the geometric time of physicist. One should be able to resolve this logical flaw and this requires that the two times are different - something which of course even child knows! If we have two times we have also two independent causalities: the causality of field equations and that of "free will". This would be the first step towards the solution.
Sean Carroll also represents what he calls consequent argument. The argument begins with an innocent looking statement that our past is fixed. Therefore free will obeying field equations is impossible since it would change both future and past. Wrong again: the assumption about fixed past in the geometric sense need not be true. About subjective past it is. Already Wheeler was led to propose that in state function reduction the geometric past changes: see for the Wheeler's delayed choice experiment. Maybe Wheeler's general relativistic background helped him to make this conceptual leap, which leads very near to TGD view about quantum jump.
In TGD framework quantum states are superpositions of classical histories and quantum jumps replace them with new ones and the average geometric past also changes. The finding of Libet that in volitional act neural activity begins a fraction of second before the conscious decisions supports the idea that we are replacing our geometric past with a new one all the subjective time.
Sean Carroll notices also the ethical aspect of problem. If we really believe that free will is illusion, we have no justification for moral rules. The criminal has been doomed to perform his crime at the moment of Big Bang and we cannot therefore accuse him.
Of course, there could be something badly wrong in the brain of the mass murderer and it has indeed become clear that our behavior correlates strongly with biology. This does not however mean that free choices are not possible. Braid disorer only changes the probabilities of different outcomes of the choices. We have the experience of free will as every reader can testify. This we must accept and try to understand the physical correlates of this experience irrespective of whether the free will is real or not.
In fact, neuroscience has led to quite concrete progress in the understanding of the correlations between biology and behavior. This has also practical consequences. Many mass murderers have been victims of child abuse or have suffered from brain tumor. This does not mean that we should allow mass murders to continue with their rare hobby. We can however do our best to prevent child abuse. Also the degeneration of some regions of frontal lobes can lead to highly asocial behaviors when stimuli usually inhibited are not inhibited anymore. One could say that there are competing free wills using the same biological body and the one wanting to perform the mass murder wins.
These issues were discussed already at times of Dostojevski and Turgeniev. The fashionable thinking was that we are nothing but physiology and that we can indeed forget the rules of moral. The people propagating this view and trying to live according to this philosophy were known as nihilists: they were mad but fully logical in their madness. Many people calling themselves skeptics today are surprisingly near to these views. Thanks to God, most of us are to not too strict in their logics and follow their conscience rather than materialistic dogmas.
Scott Aaronson's view
G. Musser has summarized computer scientist Scott Aaronson's talk about free will.
Also Scott Aaronson studies the idea of reducing free will to behavior observed from outside. Aaronson's thought experiment considers a Turing like test allowing to decide whether you have free will. A computer model of you is built using all available data about the initial state of your brain: this of course assumes determinism or at least quantum statistical determinism. If the computer is able to mimic your behavior faithfully, one can say that you have no free will. The proponent of effective free will might say that the longer the needed computer code is, the more you have effective free will. This kind of free-will-meter is of course not possible in practice except with some accuracy so that the whole thing reduces to mere mimicry, kind of parameter fit.
Aaronson represents the non-cloning theorem of quantum theory as a first principle objection against Turing test of free-will-meter. Even in principle it is not possible to construct complete copy of brain state to make a complete simulation possible. This kind of machine would be successful in what Aaronson calls Toddler test but this would be a fake success. Any toddler says completely predictably "No" to any question. We however know that the toddler expresses by behaving irrationally that he/she has discovered his/her free will (but can this kind of free will be effective?)!
Aaronson brings in special relativity and notices that free will means also backward causation if it is to be consistent with the causality of field equations. From this it would be only a short step to the realization that the causality of free will could act in the space of quantum states defined as superposition of solutions of classical field equations consistent with holography in the sense that 3-D section determines the entire space - at least below certain scale! The problem would have been solved! Scott makes a near miss!
To sum up, Aaronson dimly realizes that in general relativity - and in any 4-D Universe obeying general coordinate invariance - we live in a kind of block world consisting of 4-D blocks but the other writers continue in the good old Newtonian style. In TGD zero energy ontology would realize blocks as causal diamonds and would extend free will from a mere choice between given alternatives to creation of new worlds. Sabine Hossenfelder realizes that emergence is self deception: I cannot but agree. Sean Carroll grasps the full meaning of the absence of free will at the level of moral issues. Eagleman describes real life situations, which should be highly valuable for any-one proposing in earnest a theory of consciousness. Also the lecture of Eagleman about perceived time was excellent. To me it seems that physicists and (quantum) computer scientists should be able to forget for a moment their formulas and rhetorics making possible to get rid of a problems they cannot solve, and open their minds for the problem to get settled.