Diigo Links

Sunday, December 06, 2009

Can motivation for intentional actions have names? OR, how I answer my 4yr old's questions "If there are no people, like if they all died, everyone in the world, would Santa still be alive?" and "What was before the first dinosaur?"

I'm interested in two things here. One of them is not linguistics. The two things are applying Daniel Dennett's "intentional stance" or "intentional systems" at an atomic level to re-image the way we consider calculations of things like physics and chemistry. The second thing is then assembling clusters of intentional systems to attain something of a self catalysing homeostatic system (i.e. life). I put the origin of life at the primordial sea-floor vents, so I'm making assumptions of the required chemistry, physics, etc. based on those conditions.

A caveat I'd like to make, when using the word "atomic" I'm more interested in the "essential" meaning of the word than the typical physics it implies. It's unfortunate that atomic has come to mean radioactive, when at its basis it just means "the smallest or most elemental form of something that can still be considered something". It's also unfortunate that the idea of an atom having a definite boundary or solid shell is so popularly held. I don't want you or anybody to put walls around your atoms. Instead, think of atoms as a localized increased probability of energetic reactions.

Just this consideration of the word "atom" shows that the human race has reached a certain impasse with regards to the pop/folk world-view. We stand straddling Newton's classical view (for every action there is an opposite and equal reaction) and Einstein's relativism ( all measurement depends upon the frame of reference). On the one hand we expect the world to act something like a computer, and we expect it to be computable. On the other hand we know there are aspects of life which are bound to the frame of reference or point of view.

Some clear examples of the conflict come in climate modelling, economic modelling and evolutionary theory. None of these sciences provide classical answers, and there's a reason for that. All three of these sciences deal with highly complex systems. You can think of "complexity" as being a problem with many variables. The simple truth about complex problems is that they are not reasonably "computable". They are described as NP-complete. http://en.wikipedia.org/wiki/NP-complete

Any complex problem put to computational models is typically simplified in some/many way(s) to make it conform to the limits of computation. But this begs the question: if these problems could be solved in a non-complex way, why weren't they. In other words: The essence of solving complex problems is in their very complexity.

The natural world solves complex problems continuously and consistently. The chaotic white-water of a wave breaking on the shore, the turbulence you feel when flying near a storm, the simple moment a mountainside transitions from serene to avalanche, are all examples of highly complex interactions being resolved en mass at a nearly atomic level. There's no evident calculation time. There's no measurable communication of any historical or future intention. The entire problem is resolved continuously in the very nick of time.

One might then reason that there's obviously some very economical process of calculation at work here. Apply Occam's Razor to the idea and you come back to Newton. In the breaking wave we're seeing nothing more than very many individual equal and opposite reactions taking place. If we had enough information about initial conditions, we'd be able to calculate the position and vector of each molecule of salt-water indefinitely into the future. But when this turns out to not be true, we're left looking for the next most simple answer. Calculating turbulence can be done, but not predicatively. The precision of the calculation fails the classical test of "is/isn't", and quickly becomes a probability. Similarly the weather prediction for tomorrow is more accurate than the one for 5-days-from-now. Have a look at the Travelling Salesman Problem, or any of the many other NP-complete problems for more.

What could that next most simple explanation be? And why is it not a linguistic problem?

So having somewhat disposed of classical computation (the ability of arithmetic to solve all problems), where does that leave us? It does not leave us where the ancient Greeks left off. But in excluding the power of math from our tool belt, our situation does share quite a bit with that of the Greeks. We're left without any agreed language, process, framework, or tools to specifically make sense of the world. We're left back at the doorstep of philosophy.

Did philosophy predict mathematics? No, because at the time of first invention there was either no separation between the two, or no relation drawn. (That's how our situation differs from the ancient Greeks; we've drawn some lines in the sand) Can philosophy be expected to bring forward the next step beyond math? As much as any other discipline. But surely math is different to philosophy because math is not "linguistic", math is "real" - that there are two apples and two oranges are linguistic distinctions of class and object. But that the two apples and two oranges share a thing that is "two-ness" is "real", right? Mathematicians would have you think so. But as Albert Einstein said: "as far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality." There is a space between math and reality. It's in that space that philosophy plays.

I feel it's important to highlight the linguistic (read: anthropocentric) nature of math, the philosophical underpinnings of math, and the limitations of math because the phrase "intentional stance" seems to impart some sort of language to the things to which it is applied. In most people's minds, there's no problem with calling something the "second orange" or measuring "40 degrees". There seems to be something more "real" about attributing a number to an orange than attributing an aspiration to an orange. But I'm interested in fleshing out the subtle distinction between "intentional stance" at a base level, and "reason" or "physics". I'm interested in both pointing out the degree of anthropocentricity of attributing a number value to an object, as well as opening the box on attributing a non-anthropocentric, non-number value to an object.

The idea of attributing a desire to an atom causes an immediate giggle. Atoms can't want. They can't think. They have no will. I'm happy to agree with that reaction. I'm happy to reserve "will" for thinking animals. But I'm asking for words that we can combine like Lego blocks to eventuate in living animal desire. The idea is to impart an elemental "will" which, much as the elemental atom, combines to form molecules of will, and eventually comes somewhat alive as a group of organized intention.

Now let me split arguments into two camps. One camps says that this elemental quality must be an attribute already accounted for, like charge. The other camp would claim that it's an as-yet-unobserved attribute. To either of these contentions I'd nominate units of entropy, or a relationship with units of entropy. I have a hunch that this is where the value resides. A hunch is not very scientific, but then I am not a scientist. At least there is a measurable (Joules per Kelvin) amount of something to hang 'yer hat on there. But as hinted, a relationship with units of entropy might need to give the amount an orientation or some other dimension in order to further distinguish "want" in a valuable way (in a way that would give predictive abilities).

I don't want to be the one pushing for adding extra dimensions to a problem just for the sake of the convenience of making it solvable, just as I don't enjoy seeing people remove complexity from a problem in an attempt to make it solvable. But I don't think we can ignore the clear gap that remains between what exists and what's been explained.

Scientists have looked at individual nodes of DNA (primary structure). They expected to find all the answers, but what they found was that the secondary structure, the double-helix could not be ignored. And then a very clever team found clear indication that the tertiary structure was being used by some animals, inside the light receptors in the eye, to amplify light. This is to say that the folded shape of the double helix was different inside these cells in such a way that was advantageous to the total animal. How was that "want" communicated down? You could quite correctly argue variously that a signal is required and that a signal is not required. And depending on your classification system, you can either see that signal or not.

Where does this emergence live? When does randomly clicking Lego blocks together become a model of a house? If we can agree that Lego blocks form a group "they". And if we can finally agree that Lego blocks want to be clicked together (design stance), we could also agree that they want to make two classes of things; one class is things which have a clear utility, the other class is things for which a use is not clear. If we then observe a "user", one (or many) who then employs the utility of a made Lego structure, and thereby allows it to persist, we can start to see how the underpinnings of life may have started in the deep blue sea.

At a chemical level, no distinction is made between a primary, secondary, tertiary, quaternary, etc. structure. Distinctions are made based on difference without regard to class. Whereas class distinction has served a purpose in the progress of human understanding, these borders also serve to limit our ability to understand the dimensions left undefined. A great deal of value does reside in the individual objects which we define, but there is clearly more. One of the uses of employing Dennett's "Intentional stance" is to establish conditional relationships. To define a relationship is to interrogate that space between defined objects. Having defined many objects, it's up to us to define the parameters of the possible relationships.

I'm not suggesting that we're missing some essential "soul" dimension of these relationships. But I'm not convinced we're done considering the currently defined attributes absent the typical constraints demanded by the individual disciplines (physics, biology, chemistry). It's worth remembering that those specific disciplines exist as much as Santa Claus. If we stop believing in them, they cease to exist. I agree the demarcating borders of these disciplines have been valuable. But I think it's also very important re-draw the lines to include the newly possible.

I don't see a groundswell of support for the "intentional stance" in the sciences (not as much as memes anyway). This doesn't make it a bad idea. And this does not make it wrong. In the language of memes I think it just makes it dormant. And to my mind it's a sleeping giant.

In summary, before the dinosaurs there was goop which fell into catalytic relationships driven by the potential energy present at hot water vents in the deep primordial seas. And if there were not people Santa would not "die" because he is magically sustained by our belief ( a powerful self-reinforcing meme which lives in the relationship with parents' desire to see their children happy), but would instead merely cease to exist.

No comments: