Religious people sometimes invoke a deity as the cause of things they can't explain - a 'God of the gaps'. Scientists do something similar when they come across something that baffles them. They invoke randomness as the cause of the phenomenon. Randomness can be seen as science's 'God of the gaps'.
Both agents - God and randomness - are presumed by their respective supporters to be the reason that certain features of the world are as they are. The difference between them is that God is supposed to act with intention but randomness does not. God's degree of intention might be assigned a 1 i.e., perfect. Randomness' degree of intention might be assigned a 0 i.e., zero intention.
I do not agree that either entity - God or randomness - should be credited with causative power in this way. Scientists should not invoke deities, randomness, or anything else of a mystical nature. Nevertheless, conventional scientific thinking is that the scientific 'God of the gaps' really does exist. But many scientists will agree with me that causes inexplicable in principle should be kept out of science. When confronted by something we cannot explain, we shouldn't wheel-in randomness, because religious people can just as conveniently, and with equal justification, wheel-in God. Random causes should be kept out of science.
Let us consider Boltzmann's idea that the whole world around us could be a chance fluctuation of a gas of particles. Boltzmann suggested that the origin of our world could be like all the air molecules in a room accidentally finding themselves in a the top left-hand cubic centimetre. It is indeed possible that the molecules would do that, if they jostle long enough. For the world as a whole, it's possible that random jostling of a bunch of featureless particles caused them to fall into the special position that has led to the world being exactly as we see it today. They would have to jostle a long time, of course, but they have a long time to do it.
That would be a poor explanation of the world. Anything can be explained away if we allow that sort of explanation. Anyway, 'random cause' would appear to be an oxymoron. The very notion of 'cause' suggests that it can't be random.
An intentionality of 0 indicates that randomness doesn't care what outcome it produces i.e., it doesn't have free will in any degree. We have invented randomness to be our default causal agent, and since it is our invention, we have the freedom to give it whatever characteristics we want. We give it the ability to 'produce' things, to act in its own right and to cause outcomes, but we don't give it the slightest amount of free will. Free will is something we reserve totally to ourselves.
Our flirting with randomness as a causative agent is either wishful thinking or superstition. We want phenomena to have a cause that doesn't involve us so when all else fails, we invent randomness to do the job. For myself, I prefer not to invoke randomness but to admit that there are things we can't currently explain.
I agree with scientists that randomness can be a useful tool. Statistical tools are very useful when dealing with a large number of particles or events. A deterministic description of all the molecules in a roomful of air, for instance, would be impossible. Statistics should be used in such cases, but that's where it should stop. Randomness should never be given causative powers.
However, many people think that randomness does indeed have causative powers in quantum theory. Let us consider the reality that appears when a quantum particle is measured. Physicists say that the reality observed to appear is not caused by an underlying reality but is the result of chance. I can agree with them that there is no underlying reality. (This has been proved beyond doubt.) But I don't agree that we should invoke chance. Just because an underlying reality is not the cause doesn't mean that chance is. This is one of the most common errors of logic. We suppose that not-A implies B. Religious people could with equal validity say that God rather than chance is the cause of the outcome measured by the scientists, as that explanation is just as readily implied by the lack of a real cause.
It is possible that a quantum measurement result does indeed have a cause but we cannot know what it is. This would happen, for instance, if the cause came from the future. In that case we would not see it as a cause in the conventional sense. The appropriate response would be to revise our notion of 'seeing', perhaps changing our idea as to what a 'cause' is. A new explanation of the world might then emerge. We will not find such an explanation if we dogmatically insist that things are explained by randomness (or by God).
In the Many Worlds theory of quantum physics, each of the worlds is entirely deterministic and remains so regardless of observers. When we make a quantum measurement and get a random result, the Many Worlds explanation is that a huge number of different deterministic results occurs, but each one is realised in a different world. Different observers see the different outcomes. Since all of the worlds are orthogonal to each other, none of the observers can know what the outcome is in any world other than their own. The point is that all these worlds are deterministic. Randomness is a concept that none of the worlds has need of.
Like the Many Worlds theory of quantum physics, Einstein's theory of relativity is deterministic too. With relativity theory and the Many Worlds theory, we have two major explanations of the world completely deterministic (i.e., absolutely devoid of randomness).
That being the case, why do people insist on randomness being real? I suspect that the scientists are viewing the Many Worlds from an outside platform. They are looking down from on high and seeing the different outcomes occurring in different worlds, explaining what they see across all the worlds in terms of randomness. But it is their seeing that generates this randomness. If they were to give up their mystical platform outside the ensemble of worlds and return to the reality of one world, the randomness would disappear.
(In the Copenhagen interpretation of quantum physics, the randomness that appears in a quantum measurement is explicitly acknowledged to be caused by the act of observation.)
For a period of about ten years in the last quarter of the twentieth century, the phenomenon of chaos occupied the attention of many scientific minds. Some of those minds - those in the Ilya Prigogine camp, perhaps - thought that chaos theory got rid of determinism. They agreed that determinism continued to reside in the microscopic details but they insisted that the exquisite sensitivity of chaotic processes to initial conditions made the determinism useless. In their view, we should dump determinism and build our worldview on randomness and the 'order' that is 'seen' to 'emerge' from it.
My own view is entirely opposite. I believe that chaos theory strengthens the grip of determinism. No matter how random a process might look to be, chaos theory says that it might still be the result of straight-forward determinism. Before chaos theory, many thought that the world's obvious randomness could only have come from some primitive source such as quantum theory. Now we know that is incorrect. The discovery of chaos theory is actually another reason for getting rid of randomness. It's determinism that we should keep.
The questionable status of randomness appears also in mathematics. Let us consider the number 01100010101001010101110101. It appears random, but the mathematician Gregory Chaitin has proved that randomness can never be proved in respect of numbers. No matter how a number might look, it is still possible that it was determined by some non-random process i.e., has some non-random reality. Insofar as the world's reality is mathematical, Chaitin's remarkable result concerning numbers suggests that the existence of randomness cannot be proved in the world either.
If relativity theory, quantum physics, chaos theory, and mathematics suggest that randomness isn't real, we ought to accept that it isn't. We should let 'chance' join phlogiston, the life-force, the lumeniferous ether, protoplasm, swerve, and other hypotheses, as things that can't be proved to exist and that science fundamentally has no need of.
We would then be left with the (strongly supported) idea that the randomness we see in the world is an artefact of our observation. I suspect that most physicists secretly accept that this is the case. But not biologists. The latter require randomness to be absolutely real for their theory of evolution. If chance were not real, it could not be the cause of evolution's variations. Worse, if the randomness were replaced by man's observation, there would be no way that man could have evolved via evolutionary theory. Biologists require randomness to be real (i.e., independent of man) because if it's a product of man's observation of the world, it's impossible for man to have arisen by random means.
How do evolutionary biologists handle the evidence of the world's determinism? By not thinking about it too much. Their evolutionary theory requires randomness and that's all there is to say. If the physicists cannot prove randomness, they should try harder...
An example of this unprofessional approach towards physics comes from the biologist Richard Dawkins. He says in one of his books: 'I listed three respects in which mutation is not random: it is induced by X-rays, etc ...' To a physicist, X-rays are the quintessential means of obtaining randomness, yet here Dawkins is using X-radiation as an example of non-randomness. Dawkins does justify his position somewhat, but to my physical point of view, his whole approach to evolution does not attempt to come to terms with the determinism of modern physics. (He even attacks 'determinism' itself, calling it a 'pretentious jargon word' that he 'didn't want to burden the reader with'.) [Ref. 20] This incompatibility between physicists' and evolutionary biologists' view of the world is a serious problem for science. To my mind, the fault lies with the biologists.
I hinted above that an event might have a peculiar type of 'cause' that acts from the future. This would go against the normal direction of cause-and-effect and we would be blind to it; i.e., we would not automatically see it as a cause. Nevertheless, the future event could be the logical reason for the past event and thus its correct explanation. If something is universally believed to be the cause of something else, we ought to accept it as the cause, even when it offends our prejudices. Although we might dogmatically assert that causes shall act from the past, many scientists would recognise that the freedom to make dogmatic assertions is also the freedom to change our mind about how causes shall act.
Here's how a cause acting from the future might work. Earlier I described a computer simulation of evolution where the number 10 entered into a computer produces a kiwi bird on the screen after six hours. The computer simulation is deterministic in that the kiwi bird always appears at this time, provided the computer is initialised with the number 10. Perhaps worried by this determinism, some evolutionary researchers go to a lot of trouble to have the computer initialised by a random process, such as the radioactive disintegration of an atom. (The researchers might also inject randomness at several points during the experiment. The argument I am making here doesn't change if they do that. All these moments of randomness can be referred to the beginning, the various random numbers combined into one initialising number which is entered at the start of the experiment. In the experiment I am describing, this single number is '10', but it could be a large number.)
Let us consider an experiment where we do not know what the initialising number is. Perhaps we don't know it because the number is generated by a radioactive atom directly coupled to a port on the computer. We run the experiment and monitor the screen. After 7 hours we see a very good likeness of a 'pukeko' bird in the display. (The pukeko is a flightless New Zealand native bird rather like the kiwi but much more common.) Examining the computer we find that the initialising number in the case of the pukeko was 11, although we didn't know that before.
Now comes the important bit. Reality starts with our seeing something interesting on the screen. Until we do that, everything is random. (i.e., the default state for both input and output is randomness.) Reality - non-randomness - starts when we see something interesting on the screen - the pukeko. If the pukeko is a real object of our (simulated) world, and therefore a source of reality, it follows that the input number is required to have been 11. This is a momentous conclusion. It means that the atom's pattern of disintegration wasn't 'random', but was required to produce the number 11. (It is possible that numbers other than 11 will also result in a pukeko at the required time. The point is that a random number won't do this.)
To repeat, reality starts with us seeing something on the screen. Since the computer process is deterministic, the object on the screen puts constraints on the value of the computer's initialisation number. This number was set at an earlier time and was determined by the disintegration of a radioactive atom. If that number is required to be a particular number, then the atom is required to disintegrate in a particular way. That is the only logical conclusion we can come to, and it is difficult to disprove. We could disprove it by finding some other explanation of the atom disintegrating in the way that it did, but scientists agree that that is not possible. [Ref. 21]
We are left with our observation of an object in the present being the cause of an object in the past. This is not too different from the 'anthropic principle', where our existence in the present is the cause of our world having certain characteristics in the past.
An interesting feature of this backward causation is that time drops out of the equation, the 7 hours between the input event and the screen event becoming irrelevant. It seems that our observation of an object in the present is all that matters. We need to know the number 11 (the input number produced by the random source) only if we insist on repeatability. To learn that number, we need to know 'how far back' in the computer's processing to look, and so we need a knowledge of time. Other than for purposes of repeatability, time is not important. Even the processing undertaken by the computer turns out to be irrelevant. The only thing that matters is what we see in the present. Everything apart from the present seems to be our pathetic attempt to make a repeatable 'reality' the 'cause' of what we see...
In the real world (as opposed to a simulation of it) we see objects to exist. These objects necessitate a certain past to have existed. Equivalently, they cause a past to exist that is compatible with them. Everything in the present need to be compatible with everything else in the present, of course, but also the past needs to be compatible with the present. This compatibility requirement - our insistence on a past and present that are 'believed' - removes the determining role of time. It's our beliefs that determine things. The mysterious independent sequence we call time is our invention, something we make up to agree with the reality we have determined shall exist.
The dropping-out of time assists in the free will creation of objects. The 'time machine' I referred to earlier, which we use to travel back to the big bang to tweak the initial conditions, doesn't have to travel very far at all. It just needs to visit some part of the past that is linked to the present we now want to exist. (The past event must not have been visited before, i.e., it must not already be set in concrete.) The past event might be just a picosecond in the past - not 14 billion years. In the computer example above, the past event that we visited was the radioactive disintegration of a particular atom. That atom had a link to the pukeko's appearance and was free to disintegrate in the required way.
Although time drops out of the equation, the idea that free will is a time journey back to the big bang can be retained. It is equivalent to accepting that the big bang wasn't completed 14 billion years, but is still going on today. The radioactive disintegration of an atom in the world today is a manifestation of the big bang still happening.
The link between free will and quantum randomness is an old one, but not in the form here. The old idea was that a random world replaced a deterministic world, and so got us out of the deterministic straitjacket and allowed free will. Now we realise that to allow free will the world actually needs to be deterministic. It wouldn't be any good travelling back to the big bang to set the initial conditions if, during the 14 billion-year return journey, the world ran off the rails! No, the world needs to be deterministic. Fortunately, the scientific evidence suggests that it is. At least, the evidence suggests that the world since the big bang is deterministic. With the big bang still happening, we have a world that is fully deterministic.
Randomness might not exist in the deterministic world but the appearance of randomness certainly does. We need to explain why. We want to know why so much of our world looks governed by chance. Let us return to the apple on the table viewed by two people. As I said, each person believes the apple to be a certain object and it is the sum of the two beliefs that represents the apple to the world. (I pictured this sum as a vector resultant.) The apple enters the world of reality according to what the two people together believe it to be. This process guarantees that just one apple enters the world in spite of there being two beliefs as to what it is. Randomness enters the scene as follows. The beliefs of the two people will be (slightly) different from each other and it is this difference that surfaces elsewhere in the world as randomness.
Each person will construct a past that is compatible with the object they believe to exist. If one of them sees the apple as 'red', he or she will require redness-producing events in the past that lead to a red apple now. The other person might see the apple as orange. That will require slightly different events in the past. These incompatibilities residing in the past add up to randomness. That's all there is to it. (It is a good thing that these incompatibilities produce randomness because the alternative is logical contradiction.)
It follows that if there were just one person in the world, there would not be any randomness. This is obviously correct. There wouldn't be any reason for that person to construct a past that was random instead of a past that was deterministic. It also follows that a 'perfect' world - a world without randomness - is only compatible with the existence of one person. A person is defined by their beliefs. If two people believe the same, there are not two people but one. To have the 'same' world believed to exist - a world without randomness - everybody needs to have the same beliefs, but 'same beliefs' implies just one person.
But we also need to keep in mind that there can be no 'real' world if there is only one person. With only one person, it's not possible for him or her to distinguish between a possible object and one that is real.
All this adds up to two possible initial states for the world: (i) one person, no real world, and no possibility of a real world; or (ii) two people, no real world necessarily, but the possibility of a real world. In the latter case, the two people have the freedom to create any real world that they might want. The freedom to do this - i.e., the absence of a real world - represents a state of perfect symmetry. To create a real world, the two people have to engage in an act of 'symmetry-breaking', equivalent to choosing one world out of the infinite possibilities. Although they have no real reason to make such an act of symmetry-breaking, they might still do it. They might be anxious to provide a 'real' cause of their otherwise-unexplained existence.
This mainly completes the explanation of free will that I set out to do. On the next page I will provide a kind of summary of the ideas that have been presented. This will be followed by a page detailing the scientific problems solved under this theory.
Back to beginning of Page 1 ('A solution to the problem of free will')