Doubting Didymus

And what, exactly, is your problem with the poker deck analogy? Random shuffling, filtered by selection of good cards and discarding the worthless ones, allows a perfect hand to be generated. So evolution is the non-random selection of randomly varying replicators. What is your problem with the selection part, that you dismiss it's non-random nature so flippantly?

Albert Cipriani

Doubting,
What we have here is a failure to communicate. And it ain't mine.

I have no problem with HW’s poker deck analogy the way you’ve articulated it:
You see: there's random AND non-random processes. Duh. Now compare your statement to HW’s that I have been arguing against:
Evolution DOES depend on chance, just as much as a car depends on gasoline. Chance, or your “randomly varying replicators,” is the animating principle of evolution. Without genetic randomness, there would be nothing for the non-random selection process to select!

Why is this so difficult for you guys? I’d much rather be dialoguing with Alix and Baloo about the deeper metaphysical implications of indeterminacy. But my analysis and challenge to them is now buried under all this trivia. – Disgusted, Albert the Traditional Catholic

Principia

Albert,
Is your variation on the word "discreetly" a result of chance? How would you know?

For your own amusement (since you seem so bored): a FAQ on the notion of chance and evolution.

Doubting Didymus

Welcome to the world of scientific investigation, Albert. Remember to thank it next time you're in that gasoline powered car you keep talking about. And give me a shout when the deeper metaphisical implications of indeterminacy put food on your table, cure you of an ailment, or tell you anything at all factual about what you are and your place in the universe.

In the meantime, we trivial science types will get on with discovering the verified empirical truth about the world and everything in it.

Happy Wonderers point is that, even if randomness is the engine of evolution as we know it, it is not theoretically required for evolution to occur. That is, the mutations could be non-random and occur in some kind of completely predictable repeating cycle, and evolution and its accompanying common descent would still arise. However, for evolution as we see it today, you are correct that chance is a vital component. It doesn't HAVE to be, but it is.

But, you have attempted to argue that these two components are somehow distinct from one another. Do you recognize this voice?

Yet you appear to understand that natural selection and random variation are both equally neccesary in the neo-darwinian synthesis. Without selection, genetic information would deteriorate into noise. Without variation, selction has nothing to select. These two processes are irreversibly joined, and most importantly, the end product is not randomly determined. It is quite predictable what evolution will produce: an organism that is suited to its environment. When we expose bacteria to antibiotics, we know what trait the bacteria will eventually evolve. It will evolve antibacterial protection every time. Thus it is impossible to say that evolution (variation plus selection), is random, as its results are anything but. See?

- exhibiting patience and restraint, Didymus the Inquiring Skeptic.

Lobstrosity

Actually, you were discussing with Baloo and me (or at least I was the one who responded to your post). Alix merely made a comment with regard to my post to you and somehow he gets all the credit????

Lobstrosity

Well yes and no. I don't view it as completely quite that simple. An organism's environment is highly complex and as such there is a near continuum of ways in which it might be "suited" for that environment. I don't think the "fitness function" for any organism will in practice have one true clear global maximum. I think in general you find that there are quite a few evolutionary paths an organism can take that all seem equally "good" with regards to survival. Without a clear-cut optimal solution, the random aspect of evolution can lead to highly unpredictable results. In hindsight we can see why a specific evolved trait is beneficial and how it aided in survival, but I don't know if anyone could have predicted beforehand that that exact trait was about to evolve.

In short, what I would say is that naturally selection will always select in a non-random manner, but because it's selecting from randomly generated phenotypes and because there's no one optimal solution, the results will have a degree of randomness themselves. Specifically, I would hypothesize that if you take two isolated, identical populations of organisms in two identical environments (meaning that the environments can change together with time but will always both be the same at any given instant in time) and left them for a few million years, I think you would find two different species when you checked back.

Albert Cipriani

Dear Lobstrosity,
Sorry for the oversight. I wash my hands of Alix.

Excellent post. I understand and agree with everything you wrote. To sum up, evolution cannot escape from non-random processes. Randomness is intrinsically a part of evolution. – Sincerely, Albert the Traditional Catholic

Baloo

Albert,

Sorry about the delay - and, to give credit where due, I actually didn't respond to your query a week ago because Lob said exactly what I woud have said, only better (nice post, btw).

However, you'll notice that there really is no conflict of interest here - my statement was limited to "if only that which can be known about an isotope in the present is known, it's future cannot be predicted." This leads to a simple escape-hatch whereby any theist can simply include "knowledge of the future" in the realm of omniscience, and thus an omniscient deity is again back in the drivers' seat. But it seems that you've alread come to that conclusion, so before I say anymore, let me ask: have we reached common ground on the subject? I mean, I think we have a deal: you and I both agree that an omniscient deity, if defined to have complete knowledge of past, present, and future, would know the future. In fact, please note how I qualified what I wrote in my original post (emphasis added):
"God created a universe where not even He, with full knowledge of only the present and past, could predict the future of a single isotope."

Alix Nenuphar

And so I am to be deprived of all the fun because Lobstrosity actually did the work? What kind of message board is this!

The unknowability of the universe, as you term it, is certainly not discordant with an omniscient God - although I would quarrel with your concept of God having 'experienced' the universe.

The process by which any diety 'knows' the universe has never actually been specified by any faith of which I am aware - are you proposing a specific model?

From a Judeo-Christian point of view, you could argue that God experienced the universe during the period of the incarnation - but what, after all, does it mean for an omniscient diety to 'experience' something? Experience implies the acquisition of knowledge by the accumulation and integration of empirical data; certainly THAT would be contrary to the idea of omniscience?

Oolon Colluphid

Yes. You’ve got it. Randomness is intrinsically part of evolution as it is found in nature (though as pointed out, it doesn’t have to be).
BUT IT IS NOT THE INTERESTING PART.
It is not the part that produces complexity.

Let me repeat that slowly for you.

Random.

Variation.

Is.

Not.

What.

Produces.

Complexity.

Left to randomness, what you get is a mess. A million monkeys typing till the end of the universe might produce Hamlet. But they’ll produce one hell of a lot of crap along the way. And once having written Hamlet, they’ll continue producing crap again. It is selection (of heritable variation) that is the key.

Selection is what picks a path through the random variations, keeping at each stage only what works, and any improvements. The improvements will inevitably spread, out-competing the un-improved rivals, till they become the standard. Then repeat. And repeat. And repeat.

It is the keeping-only-the-useful-variations element that gets to the workable combinations.

Suppose you want a particular six-digit number, 123456 say. So you get a random-number generator, and set it going. The chances of getting the number are 1 in 1,000,000 (10^6, that is, 10 x 10x 10 x 10 x 10 x 10, one in ten for each number). You might get it first time, but the chances are you’ll take a lot of goes.

Suppose, instead, your generator first go throws up 249853. Then 617851. Then 900362. Then 525749.

Ah, but wait. That last one has the ‘2’ in the right place. So keep it. Now you need vary only five digits: X2XXXX. Repeat, and repeat. After a while, you’ll get something like X2XXX6. Now you need vary just four digits. And so on.

The chances of hitting on the whole number is one in a million, but the chances of each digit being correct is only one in ten.

Suppose it takes twenty tries for each digit to turn up right (which is rather an over-estimate of course). Even at that unlucky rate, it should take only 120 (6 digits x 20 goes) attempts to get the correct sequence.

Compare:

Anything up to 1,000,000 -- or even more -- by pure luck

120 if you’re unlucky.

Keeping only the correct ones rapidly reduces the odds. Keeping only those random variations that ‘work’ rapidly lets you find your way to complexity. The randomness is the fuel. The engine that gets you somewhere is selection. Which by definition is the absolute opposite of randomness.

Got it now?

TTFN, DT