Chance v. Design Part 4 (Evolution)

by xeno6696 , Sonoran Desert, Saturday, July 04, 2009, 19:35 (5619 days ago) @ George Jelliss

Matt/xeno claims: "The problem here is, that to discuss biological systems and probabilities, you and I would have to be on the same page mathematically. That means you'd have to know some dynamics and chaos, and some calculus-based probability." 
> 
> Sorry Matt, but I'm sure DT and DHW have a sufficient understanding of the gambler's fallacy, and don't need to see it expressed in higher mathematical mandarin. This is just argumentum ad obfuscation.
> - Matt is fine. - When I say "same page" I mean "same conceptual level." They don't have to be able to prove theorems, but any discussion of this nature necessarily requires all participants to be able to agree to some kind of agreed-upon language. To be able to discuss stochastic models, you need to discuss simple discrete models first. KISS in action. - 
> I looked up "Monte Carlo fallacy" in wikipedia:
> 
> http://en.wikipedia.org/wiki/Gambler's_fallacy#Non-examples_of_the_fallacy&... 
> "There are many scenarios where the gambler's fallacy might superficially seem to apply but does not. When the probability of different events is not independent, the probability of future events can change based on the outcome of past events (see statistical permutation). Formally, the system is said to have memory." 
> 
> This is certainly the case with evolution of life it is a cumulative matter.
> - And I admitted as much. That should be apparent in my discussion of assumptions to consider. Your argument doesn't change the fact that basic rules of probability still apply whether they are discrete or continuous, and I want everyone to at least know what language I'm speaking. I'm not asking them to prove Chebyshev's theorem or some BS like that, only that they have enough conceptual ammo so that they can understand my argument why chance isn't simply another "god," that is an abstract and unknowable machine that spat out life. - Life is a stochastic process, meaning that the future state of a system is predictable by both its current state and one or more random element(s). I brought up the Monte Carlo fallacy because the typical reasoning for "life can't appear by chance" states that the string of events that would lead to life are so improbable that they had to be 'helped' by some outside force. This is a restatement of the Monte Carlo, in a converse form. The Monte Carlo states "I just hit 4 heads, that means the next one must be tails." The "can't be chance" fallacy states "I just hit 4 heads. That is practically impossible!" In both cases they make the same mistake, namely that each coin toss is somehow influenced by the previous coin tosses. In both cases there is an ignorance, and a die-roll or coin toss model can show exactly where the logical error exists, namely that each coin toss or die throw has equal probabilities of occurring on all given instances. - Life has memory, but this memory is best viewed as the "current state" part of a stochastic model. The random elements can thereafter still be modeled as die throws or coin tosses. - If a generation flips a coin and gains a mutation, there is five cases to study. - 1. Immediately lethal. 
2. Lethal under certain environmental conditions
3. Does Nothing
4. Beneficial under certain environmental conditions
5. Immediately beneficial - At this point you have all sorts of "coin tosses and die throws" that can model how the organism will eke out its existence. Complexity is exponential at this point and further discussion without hard math (and a computer simulation) is moot--a plain-english discussion will be a book, for just this one mutation. - This is why I said a full discussion of stochastic life models requires a level of math that at least dhw indicated he lacked.