Saturday 2 January 2016

Complexity, inevitability, and life — Evolution 2.0 on Unbelievable?

Listening to the latest Unbelievable? show from Premier Radio I was struck by what appeared to be a failure of imagination on the part of Perry Marshall, who was debating evolutionary biologist PZ Myers about the former's recent book, Evolution 2.0. Not being a biologist of any kind I'm unable to comment authoritatively on the actual mechanisms of evolution, but having followed PZ Myers' blog Pharyngula in the past (less so these days) I'm fairly confident he knows what he's talking about when it comes to his own subject. Perry Marshall's background, however, is in engineering and marketing, which on the face of it should make me wary of pronouncements that are outside his field of expertise.

Myers rubbished pretty much everything Marshall proposed, and given the above I'm prepared to accept that Myers is right and Marshall is wrong. The debate was fairly technical, but seemed to me to boil down to Marshall's claim that the “random” part of random mutation is insufficient to explain how evolution works (notwithstanding other aspects of evolution such as horizontal gene transfer).

At one point Marshall stated that the code in DNA could fit on a Compact Disc, and that if you eliminated “junk DNA” the code would be merely ten percent of what could fit on a CD. The core of his argument appeared to be disbelief that such a relatively small amount of information could produce the complexity we see in living organisms today. By comparison he cited the amount of code required to install modern computer operating systems such as Windows 10 and Mac OS X.

Marshall's engineering background has hampered his thinking here. Engineers who design systems, be they engines, bridges, or computer operating systems, need to specify mechanisms in minute detail (or make use of minutely detailed specifications already available) in order to make their systems work. This notion of "engineering ex nihilo" is what in my opinion leads to the essential failure-of-imagination exhibited by intelligent design proponents and creationists (of whom a disproportionately large number are engineers) — “it's all so complicated it must have been minutely designed by an intelligence of some kind.

But imagine a software programmer who has never encountered fractals is shown a picture of the Mandelbrot set, and is given the task of writing code to generate the same picture from scratch. Without knowledge of the simple equation that produces fractals the picture could indeed be generated, but I suspect the code would be somewhat large. Or imagine a manufacturer of breakfast cereal wants its packaging department to come up with a special gadget to ensure that each carton of cornflakes contains a gradation of flakes, such that the larger flakes are mostly towards the top of the carton and the smaller ones mostly towards the bottom. I'm sure such a gadget could be made, but it's not actually necessary as the cornflakes tend to sort themselves out this way on their own.

Such self-organisation is, in my view, an aspect of the discussion about complexity that is often overlooked. If things inevitably organise themselves in a particular way, trying to make them happen in other ways, against the natural order, will indeed require complex intervention. “Going with the flow” on the other hand, will often require no intervention at all. It seems to me that much of the complexity we see in nature is there because in a given environment, things tend to work out that way rather than any other, just like in a packet of cornflakes.

A small part of the Mandelbrot set
This is applicable in other systems too, such as how you organise your life. For instance, it makes sense to keep things you need regularly in designated places, so that you don't have to embark on a time-consuming search every time you need them. If you need to take something with you when you go out, you could set an alarm on your smartphone to remind you to pick it up at the appropriate time — or you could simply place the item where you will see it when you do go out.

To put this another way: don't expend energy trying to achieve things in spite of your environment. Rather, create, encourage and adjust your environment such that it allows those things to be achieved automatically. (There you go — who'd have thought a debate on evolution would lead to productivity advice and life-coaching?)

UPDATE 2016-01-04: Perry Marshall has published online his transcript of the debate, along with some restrospective comments:
...And here's PZ Myers' response to the comments:

UPDATE 2016-01-10: Looks like this will run and run. Perry Marshall has responded to PZ Myers' response to his comments on his transcript:

UPDATE 2016-01-12: ...and PZ Myers further responds here: 

UPDATE 2016-01-22: Will this never end? Perry Marshalls's next shot:

...And possibly the last from PZ Myers?:

PZ Myers' blogpost about his encounter with Perry Marshall is here: