Notes on Some consequences

In this text there is fundamentally nothing further developed in the model of Mind, but I have some need to put several notes on the consequences of what was said in the "Justification".

What about Quantum Mechanics ?

Let’s analyze further relation between the impossibility to bring out mathematical model of brain, and the apparent possibility of physics to model what we observe, and try to see what is about QM which gives the possibility to resolve this contradiction, and if we can try to resolve the contradiction in some other way.

First, there might be other way for addressing this contradiction, and that is by saying that there are “emergent properties”, but, I will try to show that such theories can’t give real explanation of the contradiction at hand.

Those theories claim that certain configuration of certain parts can have properties which are not deducible from those parts. In such way the parts would have a set of laws which would describe their individual behavior, and give possibility for their incorporation in larger structures, but we wouldn’t be able to deduce how those structures would work.

Seen in that way, in my view, there might be two ways to see the “emergent properties” theory…

One way to view the “emergent property” is through imagining a suppressed faculty of parts which can get on the surface just if the particles get into certain configuration. Seen in this way “emergent property” is present as a potential in each particle even it is not in the configuration which would allow it to actualize that potential.

Now… this way of viewing emergent properties doesn’t seem as stating anything fundamental about the way things are. It seems to me that it just gives us certain authorization to build theories on multiple levels (e.g. theories about separate particles, and separate theories about their configurations), but it really doesn’t say that those emergent properties aren’t caused by the properties of the particles… just that it might be more practical to analyze them on separate level. Even if they say that there are theoretical difficulties in deducing those suppressed properties, the speculation would still see them as potentials present in the particles, not somewhere outside of it. So, I wouldn’t say that this second interpretation of “emergent properties” has enough power to resolve the contradiction I’m trying to address.

Other view is where emergent properties wouldn’t come from the potential in the properties, but their “configuration” or “dynamical configuration” would somehow induce the “emergent property” from some “otherness”. But even in such interpretation the “emergent property” would be determined by the configuration/dynamics of the parts, and we couldn’t really make difference between this view and the previous one, as it presumes that same configuration/dynamics of same parts would every time result with same properties (be them considered emergent or not).

In short… if one says that “emergent properties” of a whole are those properties which can’t be deduced by the properties of its parts, and then says that those properties “emerge” in a very specific and deterministic way from the parts is obviously talking just about the lack of knowledge on his side, and not about “the whole not being fully determined by the configuration of the parts”.

So I take it here that as much as I understand what “emergent properties” should mean, they can’t be used to address the contradiction.

But is there any other possibility?

I think that the possibility is already in the statement:”Whole is configuration of parts”.

The propositions of type “A is B” are putting in relation the subject A and the predicate B, and are saying that the subject A, can be seen as B, but are not saying that A is “just” as B.

So if we say “duck is bird”, we are saying that duck is bird, but implicit in that statement is that “duck” is not just bird, that it has something more about it, which distinct it as a duck from other types of birds.

So if we take this approach to the “whole”, by saying “whole is configuration of parts”, we see that it can mean that whole is configuration of parts, but that it has something else in it.

I would try to restate it in different way to present it more clearly… I would take that we have some specific whole in mind, and we name it W1, and we agree that it is specific configuration of parts with certain dynamics – DCP1. Then the proposition would be “W1 is DCP1”.

Now having in mind subject/predicate relation, we can see that it is logically possible, that even W1 is DCP1, the relation doesn’t really say everything about W1. Or, that there can be other W2 which would be also DCP1, but there would be other things where W1 and W2 would differ. (In same way we can say “duck is bird”, and “penguin is bird”, and still “duck is same as penguin” isn’t true).

So, it is logical to say that even two wholes are same configuration/dynamics of same particles, they would have different properties.

What Quantum Mechanics gives us is the way to see this possibility which was one side of the contradiction from scientific viewpoint, which is the other side of the contradictions.

It is because QM gives us way a system which can be determined as certain dynamical configuration of particles, not to behave (give same results to measurements), as other system which can be determined as same dynamical configuration of particles.

Why QM is such, and have such properties ? To answer those questions we must develop the Mind (or give a model of Mind) which can do concepts, propositions, logics and mathematics, and finaly form physical theories.

Modeling of Artificial Intelligence

How do the conclusions I drawn here, affect the idea of engineering minds, engineering the kind of AI (artificial intelligence) which would have consciousness?

To create (construct, engineer) something we first must have in mind what kind of behavior from engineered system is needed (what is our goal). When the goals are given, we further try to build a model, first rough one which would give possibilities for those goals, and that more and more detailed which would reach those goals.

But from the conclusion that there can’t be mathematical model of the Mind (nor brain), creating AI can’t be done that way.

We can argue that lots of time engineering is not about building a full model first upon known facts which we connect in our mind – lots of things were engineered even before we knew the details of their works - And of course in one way in looking at things, we can’t create anything by ourselves, what we can do is change the “outside” form, to change the configuration of the matter or energy which is already present in our surrounding – so to require understanding in order to engineer would be requirement to get to the universal physical law first, which would be nonsense.

But in this case it doesn’t make difference, as we can’t even see possibilities of some of the goals (consciousness, unity of consciousness, feelings etc…) of any “rough” mathematical model, neither we can find them as such in the nature in order to put them in our creation.

And such model – a rough idea of a way in which we would reach the goals of our engineering must be present.

In the last group are systems which are based on competitive learning, or some kind of “evolution”. The idea is that even we are not smart enough to construct the conscious AI ourselves, we can build system in which separate “entities” would compete between each other (for resources, or for each other, i.e. predator – pray), and have potential to mutate.

But something is again wrong, even it can seem as good idea at start –This system would again be human abstraction, it would be just the engineer has put into it, namely mutations would be just abstract mutations, their possibility would cover one “mutation space” implicit by the design.

In such way this kind of system fails to be fundamentally different from other “normal” systems of conscious AI, as in fact it doesn’t really matter if one system should be conscious right away, or after several days/months/years…

It would assume that such “evolution” would work, if the beings which mutate and evolve weren’t determined as mathematical models with mathematical mutations, but I can’t see how that can be possible.

I can’t think of any “freedom” we can put in such system except the 2^nd law of thermodynamics, which can’t surely bring positive results.

But this has other implications – that if such is the case with engineered evolution, but evolution of species on Earth led to conscious intelligence, then this “natural evolution” is also evolution with mutations in the “spiritual space”, the space where we put the Mind, and it can’t be fully explainable by mathematical model also. So, to rephrase, if brain can’t be mathematically modeled, then neither can the process that brought that brain into being.Even the principle of mutations and natural selection can be the real principle of the evolution, but it is the nature of mutations and life which are seen as impossible here to be modeled.

This kind of model makes it easy for us to think of consciousness or awareness as present in other animals; and even more it would be very strange if evolution made us conscious in such a way, and made the more “undeveloped” animals intelligent in some other ("purely mechanical", whatever that would mean) way.

Such idea also brings out the question: where is the line between awareness and non-awareness in the evolution? But, really, I can’t address that question just from the conclusions I got so far, as though “spirit” as was defined maybe equals “mind” and correlates with what we see as brain, we can’t say what kind of “otherness” correlates with what we see as certain life form. What I can be sure is that each life form is more then what it seems to us (is observable and modelable by us), as it is product of mutations in space which can’t be mathematically modeled.

Previous:Justification Next:First Principles