Three Seas Forum

the archives

dusted off in read-only

  •  

Countering the Argument posted 16 January 2009 in NeuropathCountering the Argument by Thorsten, Candidate

The Argument in short

Quite a lot of the contents of 'Neuropath' revolve around the Argument, story internally made in discussions between Neil and Thomas. Simply put, the Argument states that the human brain is a deterministic biological machine processing input from its environment and generating output from this input based on deterministic processes. As a result, notions such as 'selfhood', 'free will', 'decision making' or 'meaning' and ultimately 'consciousness' are illusions. While the mind thinks it executes a plan to realize some future goal, according to the Argument, the underlying reality is that the past state is just computed forward, the seeming future goal towards the mind proceeds is simply an illusion, in reality it is the past that determines what will happen, not the future vision.

In the novel, the Argument is illustrated by fictional neuroscientist Neil demonstrating that artificially stimulating and inhibiting several functions of the brain generates the experiences of desire, will, love, personality, spiritual experience and self which can be switched on and off externally, apparently demonstrating the claim that the mind is a mere machine.

Assume for the following that neuroscience can actually do all this what is described in 'Neuropath' (contemporary neuroscience actually can't, and based on the way genetics has not delivered on initial promises but discovered that things are actually much more complicated, my guess is neuroscience will likewise discover that things are far more complicated) - is the implication really that the Argument is true?

At least in the way it is made in 'Neuropath' and the way I have seen it presented by neuroscientists, I don't think it is. In the book, it actually appears very compelling, but it's just a trick. It's really full of substantial gaps which are (quite cleverly) glossed over in the text, and in the end it's down to belief, nothing more. The gaps roughly fall into three groups. With increasing severity one finds: Hidden assumptions in the logic of the Argument, a reliance in reductionism and an application of logic and rational thought in situations where it is known to fail. But before we go into the details, let's investigate some of the notions like of free will or decisionmaking without any input from neuroscience.

The notion of free will

Let us assume for the following a situation where two alternatives A and B are given and a person has a free choice between A and B, but cannot choose both A and B, and likewise must choose either A or B. This excludes situations where the person does not care about any of the alternatives - if A is 'you get 10 dollars' and B is 'you get 10 dollars' then this is not a choice, you might as well throw a dice. Likewise, if either A or B has very negative consequences, like A 'you get 10 dollars' and B 'you get life imprisonment', the choice is hardly free (I refrain from a detailed analysis of how to define negative consequences here, I don't think it's central to the discussion).

The notion of free will states that before the choice you think you could choose A or B. After the choice, you think you could have chosen the other alternative. Thus, how would one prove or disprove the notion?

Let us begin with the classical analysis. The simple answer is - you can't. There is no experiment that could establish before if both choices exist - because both of them are in the future, and consequently neither of them exists yet. But you can't do it after the choice either as long as you require a consistent past, because if you have chosen A you have in fact not chosen B and are left with the impression that you could have, but no prove. But if you have chosen B, you are in the very same situation left to ponder if you could have gone for A. The need to have a consistent past always creates the impression that the outcome, whatever it was, was inevitable and determined, because there is only one past.

(By the way, the same is true with many references to property X of the mind being the result of an evolutional adaption to the environment. You can point at everything and claim 'It is here, this proves that it must be the result of natural selection and evolution, otherwise it would not be here.' However, evolution doesn't quite work like this.)

Why do I talk about the need for a consistent past? Because there is the quantum analysis of the problem. And this states that you do in fact choose both alternatives, with the state vectors A' (having chosen A) and B' (having chosen B) are both part of your state after the decision, with a weight somehow determined by your personality. We don't experience any of this, and I have no clue why this is so, but since the point for me here is to counter the Argument, not to prove what is real, I don't have to explain it. Anyway, in the quantum analysis, there is no consistent past in terms of events as we usually know it, the requirement seems to be one of perception only.

Back to the classical problem: We could think about simply getting the subject back to the alternative and offer a new choice to see if the person now goes for the other alternative. But that doesn't work, because the situation is not the same - the subject has now a memory of the previous choice, so we still don't know if the other alternative could have chosen.

So, let's do a gedankenexperiment. I present you the store and reset chamber in which we can store a snapshot of the world in time and recover it with a button. We test someone in this chamber, he makes a choice, we note the choice and restore the moment before the choice and let him choose again, and thanks to the amasing chamber, it is really the same situation.

What outcome would we interpret as the action of free will? If a person, being in the same situation multiple times, chooses the same alternative every time, we'd conclude that he is determined by circumstances. If however he chooses different alternatives in the same situation, we'd conclude that the choice is random. There is no outcome that could convince us that free will is acting.

What this gedankenexperiment does is to illustrate that the notion of free will has nothing to do with proving the factual existence of alternatives or with determinism vs. randomness. It has to to with factoring imagined futures resulting from A and B into the decisionmaking process. Unlike my cat, I don't always respond directly to seeing food with running towards it and eating. Instead, even when I'm hungry I can project a future in which I will have a lavish dinner and should not spoil my appetite before, which influences my decision-making process (the Argument would still state that this is an illusion, I'll counter that later). So this is why I claim to have free will, but not the cat, because I can imagine virtual alternative futures and let them influence my decisions.

But of course I am determined by my will. That is exactly how we use the word - a strong willed person is one who is determined to do something, who is very predictable in his decision-making and sticks to his ideals instead of changing plans randomly. Freedom is not having 20 alternatives open to me (out of which I still can select only one), freedom is the lack of constraints which would prevent me from doing the one thing I want to do. I can't will in any other way than I am.

Who makes the decisions?

There is the argument that neuroscience can detect a decision by brain scanning before it becomes conscious, so consciousness doesn't make the decisions.

I am not surprised - how could it? Consciousness is a state, an experience, a self-observation, a self-reflection - not an active agency. So the decisions are made outside consciousness and become conscious afterwards - if you watch yourself carefully, you can observe this. I had situations in which I knew a decision was made, it just had to wait a week until it drifted into consciousness. It's the problem of finding what my will actually is (which is at times difficult). The mistake is to think any of this implies that it's not I who makes the decisions - of course it is! The oracle in Delphi already had γνῶθι σεαυτόν 'know thyself' inscribed on the entrance. This would be quite pointless if 'self' would be identical identical to 'conscious self'.

The point of all this is that neuroscience doesn't offer anything which people haven't worked out long ago. Before one claims that it contradicts our notions of something, it's a good idea to investigate what the notion actually is. After this excurse, back to the Argument.

Hidden assumptions in the Argument

In order to expose some weaknesses in the argument, let us consider the scene in which Neil induces a spiritual experience, the perception of the presence of god, in his victim. A reborn Christian later in the book is deeply disturbed by the fact that such experience can be induced artificially, and it seems to show that there is no god and no soul.

But let's replace the experience with something harmless - assume Neill had induced the perception of an apple. I am prepared to guess that no one would conclude from the fact that you can artificially induce the perception of an apple in a person that there are no apples.

So, if you have the prior notion that apples are real, you interpret the experiment to reveal something about the nature of the perception of reality, not about reality. But if you have the prior notion that god is not real, you are tempted to interpret the experiment to reveal something about the nature of reality.

But where would the prior notion that apples are real come from, if not from prior perception of apples? But if you accept that prior perception of apples argues for their independent reality, you have to find something other than the experiment to argue against an independent reality of god or a soul if you are arguing with a person who has experienced prior perception of god. The experiment does not tell.

To give a similar example - I may observe that a person cannot walk in spite of telling me he wants to. If this person's legs are broken, no one would say that his mind is damaged, but that the means by which the mind causes motion are damaged. If the person has a spinal injury, again I haven't ever heard the claim that the mind is damaged, but that the means by which the mind causes motion are damaged. But if the brain is damaged, neurologists suddenly insist that now it must be the mind itself, and not the means by which the mind expresses itself, that is damaged. But that is of course stating a belief, not a fact.

An analogy with the hardware/software of a computer shows the potential flaw here: If my computer has a buggy memory chip, it will do funny things. There is no point in trying to attribute the problem to the software, I can reinstall the OS, I can change from Windows to Linux (which I'd recommend anyway) - nothing on the software side is broken. So when I manipulate the hardware, I will manipulate the output of the computer, but I cannot conclude that this does anything to the software. Or who really believes that trashing his laptop will kill Windows? Software is information, a structure, changing a particular realization of the software doesn't change the software. So, what if brain manipulation is just changing the hardware - of course the mind cannot run properly on damaged hardware, that's just what you see - doesn't mean the mind would be gone. I don't know if the analogy is true, but the argument simply assumes that it is not, and that's again down to belief.

Reductionism

Science is often confused with reductionism, which in essence states 'find the parts and explain the properties of the whole as properties of its parts'. Thus, explain why humans eat: Easy reductionist problem, because the human body consists of cells, cell biochemistry needs nutrients, therefore we have to eat so that the cell biochemistry runs.

Reductionism works in cases where a few causes on the smaller scale can be identified as a reason for a phenomenon at larger scale. We can deal with logic in situations where three facts imply something. It's rather different for situations in which a million facts imply something, but not individually, only taken together.

Consider a painting, the Mona Lisa for example. Find its parts - small grains of pigment on canvas. In what sense would they 'explain' the painting? The Mona Lisa arises as a larger scale structure within these pigment grains, it can't be seen from the perspective of the parts. The Argument explains everything that cannot be explained by reductionism as illusion - consciousness is not real, because it is not in the underlying function of the neurons, therefore it is an illusion. I guess it would say that the Mona Lisa is an illusion, and in a sense it is.

Well, but why stop at the neuron level (except that neuroscientists are familiar with it)? Let's go further down the scale, into the elementary particle structure. Down at this scale are only fluctuating quantum fields, quarks, gluons, electrons, photons and other fields. They 'exist' for minimal periods of time, one cannot even point to the fields which make up a proton or an atom, because they have no identity, one cannot point to a field and say 'this is vacuum background and this is proton'. There is no way to explain how these fields imply the existence of a neuron from their basic properties. So by the above argument, the neuron itself is an illusion. In reality, there is no neural machinery doing anything - there are just quantum fields extremizing the action described by their Lagrangean function.

But presumably, the elementary particles when seen from a yet smaller scale are also an illusion.

So, consciousness is no more or less real as a neuron. It is simply a phenomenon at a different scale, which is poorly understood when viewed from the wrong scale. It's reductionism which does a bad job here.

The same trap opens up in claiming that 'in reality' there are no future goals towards things develop, only past condition from which everything follows. That's simply wrong. The underlying quantum states stretch through 4-dim spacetime. They have a structure which is given by the equations of motion, which states that if you know them on any 3-hypersurface, you can compute them in the whole of 4-dim spacetime.

The way this is usually done is to specify them in the 3-dim space of 'now' and to compute them into the future. But one can specify the evolution endpoint and compute backwards, it makes no difference. One can even compute sidewards - specify the world on a 2-dim sheet throughout the whole universe at all times, and compute how the rest of 3-dim space looks like. It. Does. Not. Matter. There is a 4-dim state with a certain structure. Viewing it as causally following from the past is no more correct as viewing it as developing towards a future endpoint, it is just another way of organizing our lack of intuition for 4-dim structures.

The precious time along which things are supposed to develop is also not an independent coordinate but a dynamical entity - it can wiggle, bend, fold upon itself, terminate or interchange role with space. There is nothing 'in which' time wiggles - only the 4-dim structure. Time itself is no more real than consciousness.

Applicability of logic

Perhaps the most serious point that can be brought forward against the Argument is that it is self-defeating. I was always under the impression when people are confronted with a logical chain of reasoning which ended in proving that 'I think, therefore I am' is wrong, they would realize that something is not right in the way logic has been applied. Turns out I was wrong.

So, the Argument is based on science, scientific method in turn is based on rational thought and application of logic. As 'Neuropath' mentions, one way to counter the Argument is to argue that science is not applicable, and Bakker goes into some length arguing that this doesn't work.

Well, it does, because arguing the success of science over religion by thermonuclear explosion vs. burning bush and science being the only system which produces unpleasant truths is misleading. Logic is an excellent tool if you apply it to some outside phenomenon, but it fails miserably when folded back onto itself.

Logic is not absolute, it relies on a choice of axioms which state what a valid deduction is and what not. Mathematics doesn't define the axioms, you can have many logical systems with self-consistent axiom systems. What is usually used in science is a particular choice of axioms. For example, we usually use transitivity: If A implies B, and B implies C, then A implies C. If Socrates is Athenian, and Athenians are Greeks, then Socrates is Greek. Sounds perfectly reasonable, but we can do logic without it, we just use it because we believe in its applicability. Same with connecting true statements - if A is true, and B is true, then this implies tha (A and B) is also true. If 'Socrates is an Athenian' is true and 'Socrates is a philosopher' is also true, then 'Socrates is an Athenian philosopher' must also be true. Sounds also very reasonable.

However, let's see what happens as soon as we introduce self-referencing statements: 'This sentence has five words' is true, 'This sentence begins with 'T'' is also true, but 'This sentence has five words and begins with 'T'' is obviously not true. The axioms don't survive self-referencing statements. There is a general theorem by Goedel which states that for any formal system (such as logic) there are statements which truth or falsehood cannot be decided within the system, although some of them can be obviously decided outside the system. The core of the proof uses self-referencing.

Now, the Argument is a prime example of self-referencing, although this is never stated. If one could prove that reason, self, intuition or consciousness are just meaningless concepts created by neural machinery, then the same could be shown for logic, which is, after all only yet another function of the conscious brain. But if you could prove that logic does not work, well, there would be no Argument. Thus, if the Argument is true, it cannot be made. If it cannot be made, it's down to belief and intuition.

Funnily enough, intuition can cope with self-references far better than logic. We are able to simply see through the paradox pattern, maybe because consciousness is in its very nature a self-referencing process. So, the Argument applies logic in a situation where we know logic fails and where we know intuition to be the better tool.

What does that mean?

I don't know of course, but my interpretation is that neuroscience or Neuropath or Neil don't in any way bring us closer to how 'reality is'. Granted, they show us an alternative picture of reality, based on a different set of assumptions and beliefs than usually done, but there is no compelling reason for this to be 'true'.

Thomas would of course argue that I try to argue away an unpleasant truth. But I'd argue that the fact that a claim is unpleasant doesn't make it true, the fact that a claim is pleasant doesn't make it false, and that science is just one way to organize reality (being a scientist, I have to be careful with these statements, I know how I mean it, but it is easily misunderstood and misused...), known to be problematic in the very situation he wants it to be applicable. So if people are prone to delude themselves, what makes neuroscientists think they are the exception?

'I think, therefore I am' seems a far more useful starting point for investigating consciousness - it has the advantage that it is at least not self-defeating. view post


Countering the Argument posted 30 May 2009 in NeuropathCountering the Argument by sciborg3, Commoner

Excellent. I also wonder how unpleasant a truth the Argument would be - it would seem to me how much regret you bore on your shoulders would decide that. Some people I know would love to believe they didn't have to take responsibility for things they have done in certain situations. Others feel like their achievements don't mean as much if they're work was an illusion or predetermined.

Others feel the deterministic structure is consistent with God's plan, and thus see the brain's cause and effect as the means in which His/Her/Its/Their will(s) manifest(s). Also as you (and Eistein) state, there is no deterministic movement across time, there is no past/present/future.

Really I'm not even convinced it makes things less meaningful as under the Argument words, love, art are just fabrications of a causal chain unfolding since the Big Bang. But that means meaning is something we as a species has created, a bottom up meaning rather than a top down meaning from the Forms or God or what have you. We may be characters in a story we can't change, but we can appreciate the story we've been written into.

It would be a Sisyphean existence, to be sure, but it would still have meaning. In fact, given that the meaning comes from us rather than some apprehension of external MEANING, it might be *more* meaningful. Depends on how you look at it, and of course trying to decide which form of meaning is *more* is like asking what is worse - slavery or the Holocaust, rape or murder, etc... view post


Countering the Argument posted 03 July 2009 in NeuropathCountering the Argument by Callan S., Auditor

In terms of free will and decision making (and taking it your summing up of the arguement is roughly on target - it's been awhile since I read it), I think the utterly deterministic model is a self forfilling prophesy. Essentially the human mind (probably alot moreso than any other animal) can to a degree, observe itself. This creates a powerful feedback loop. Maybe someones addicted to cigarettes. But unlike an animal, they can see the hunger in them - they can forceably try and block it. Sure, maybe they'll cave in latter. But an animal would just go smoke a cigarette - they are that deterministic. A human looks at themselves working and their actions aren't soley on the animal level - their actions are affected by that self reflection.

Is that self reflection deterministic? Essentially yes, but it's a far more bloody complicated determinism. Because the system isn't just operating on stimuli from the outside world, it's operating on stimuli from the inner world, which is affected by the outside world, which is affected by the inner world, which is...and so on. No doubt for some things a single feedback loop could even go on for years. Bloody complicated - to just call us a deterministic machine is to indulge in a world simplifying illusion itself - like the illusion of 'solid' objects, when objects aren't solid, they are mostly empty. To see it as determinism is a simplification illusion, rather than a reason to think the human mind operates like simple clockwork. Objects appear solid. The processes of the mind appears to be clockwork. These are the illusions. view post


  •  

The Three Seas Forum archives are hosted and maintained courtesy of Jack Brown