Saturday, January 07, 2012

Tidier than thou!!

One of the most profound ideas in the book i am reading currently (Thinking, Slow and Fast by Daniel Kahneman) is that we - humans - are hardwired to think of our world to be more coherent and tidy than it really is. That got me thinking.

We are forever looking for "explanations" and "causal" connections where there might be none. We are always looking for the proverbial "why" - which is undeniably at the root of all knowledge (and much hokum as well) but is also the cause of our oversimplification of the world. I have written in an earlier blog about the human obsession with easy generalities. Kahneman also observes the same theme and attributes most of it to (surprise, surprise) evolutionary forces. I agree with him. It can be easily demonstrated in a simulation or a controlled experiment that in conditions similar to those in the early human society (or actually even in proto-human era) the individuals which are inclined to find patterns and causalities are more likely to survive than those who are clearer in their understanding of the world but also by the virtue of that remain rather confused in general. It is better from a survival and progress point of view that one creates a questionable theory and then is guided by it than being directionless because of acknowledging the inherent complexity of the world and thus the unlikelihood of a given theory being accurate.

But progress and activity aside, this tendency to generalize and rationalize has also meant that we have an over-coherent model of the world. We see patterns where none exist and we see cause and effect where chance prevails.

That leads me into another scarier direction. It is believed in general that the natural sciences like physics and chemistry have a rather solid understanding of their domain while humanities and social sciences are lot less exact. This is based on the belief that natural world "follows" order and is governed by "laws" which may not be fully known to us but are "out there". If you look deeper however you would realize two things.
1. For one, our laws of nature picture does not really explain much. As a thought experiment if you freeze the world around you for a second and try to predict what happens next (in the physical non-living world) given the current state of the world, you will realize what i am referring to. Sure you can predict that things without support will fall and broken things wont reattach by themselves. But very soon you run out of laws-driven explanation of all occurances and most things thereafter are explained as "random" (a more accurate description of this is "random symmetry breaking" - beautifully explained in the book "Theories of Everything" by John D Barrow.)
Even at the cosmic or microscopic level, the laws-explained portion is limited. Much is explained (or "explained") by randomness. Its not even that we dont know those laws yet. We have acknowledged that these are not lawful occurances. They are just random.
So yes, we are far more evolved vis-a-vis our ancestors who thought sun went around earth (which by the way is not technically inaccurate in the post-relativity world!) but we are far from knowing how the world works. At best we have a decent handle on recurring patterns in interaction within the physical world. We call these laws of nature.
2. Secondly, what we might be overgeneralizing as laws can simply be transient patterns in this specific space-time coordinate set. These might be patterns mined out of data available to us. There might not be a Plank's constant out there. It might simply be that the way things are in recent years (millenia, billions of years whatever) can be modeled as if they had a constant like that.
Which can then hold for laws of nature in general. These are our constructs - and are far more stable than anything we have seen in the rest of the world, but are simply far slower to change, maybe at this time. In another time and another location in space, even the rate of change of these might be much higher. There if some sentient beings were out with their telescopes and microscopes they might end up with an entirely different set of laws because some other patterns are constant there.

Using Occam's razor is it not more sensible to believe that there are really no "laws" but simply a spectrum of less or more stable patterns - the most stable being labled "laws"?

Is there a todo here? Probably not. This is more of a realization if you will. Not very disturbing also if you realize that it does not change anything in the material sense per se. But it is definitely a more humble approach to understanding our world than attributing fancy "laws" to it!

And generalizing from the natural sciences to all human endeavours, i would state that dogma is intellectual paralysis and end of the road for building knowledge. In this sense, Friedman while stating that "inflation everywhere and always a monetary phenomenon" is no better than 17th century Church that gagged Galileo! In the true quest for knowledge, the least we can do is remain tentative - always!!