Split Thread: Is evolution stochastic?

[jimbob]

So, random mutations can (and do) significantly affect the outcome of evolution, which includes altering the fitnes landscape for all the other organisms in the ecosystem.

How does this make it nonrandom (which is what I am arguing against).

I have said that in stable ecosystems, over moderate timescales, and with sufficiently large populations then the random aspect can be ignored, but this is only a special case.

Why do you think "nonrandom" is a beter word than "random", when "nonrandom" is incorrect, whilst "random" is merely misleading?

[cyborg]

Ugh.

Did you even read/understand what I just said?

[my_wan]

For this debate, or peoples characterizations of others claims, to make any sense whatsoever depends on what definitions are used for what.

cyborg, jimbob, etc., answer these questions.

(1) Is random and deterministic antonyms?
(2) Does randomness rule out determinism?
(3) Is your answers here model specific or do they also apply to the actual system being modeled?

Differences in how you answer these questions determines what others think you said that you didn't. This leads to accusations of false claims, lying, etc., etc. So answer them or waste your time.

[jimbob]

Well I read it as this:

Originally Posted by cyborg

Originally Posted by jimbob

So we have accepted that mutations can cause significant differences in outcomes.

Uh, I've never said anything different.

So the course of evolution is sometimes altered by random mutations, which are inherent to the system.

The fitness landscape is subject to random change. Which I now think you agree with.

Quote:

Quote:

You now say that these mutations mightn't be random.

No. I've always said that the underlying cause of mutations is irrelevant.

It's like you've managed to read a couple of hundred posts from me on this issue and you've still not managed to pick this up. I've never changed what I've said.

So are you saying that for a given stable ecosystem, that evolution is nonrandom? I would say that in this situation, given time, and assuming that no disruptive mutations ocur, then this would be the case. This is like saying that in situations where randomness can be ignored, randomness can be ignored.

In real life, the ecosystems are subject to self-imposed random changes, when random mutations alter the fitness landscape. I agree that these are not very common events, but over geological timescales (and a lot shorter) this is very important.

[articulett]

I think Sol should repost his 3 choice query, because it's clear people who answer the question the same way understand each other, and those who can't answer it or make up their own answer cannot. They will always hear people saying things they are not saying in order to prove their "point" while repeatedly missing what is said to an embarrassingly impenetrable extreme.

People who think it's informative or useful to call evolution "random" without caveat or further explanation need to converse with other "random" people living "random" lives to confirm their "random" theories of how best to convey understanding of well understood processes that they don't seem to understand as well as they imagine.

[GreyICE]

Originally Posted by my_wan

It can quiet often, for practical reasons, be only semantics. It can also be of immense theoretical value to define the deterministic underpinnings of a statistical model. The existence of the atom was established by a comparison of the non-deterministic classical thermodynamics and the deterministic statistical mechanics via Brownian motion.
http://lorentz.phl.jhu.edu/AnnusMira...s_brownian.pdf

No, it's of no theoretical value. Brownian motion is an observed phenomena. Explaining it is obviously of value. You are taking an observed phenomena and stating, as an article of faith that it is both deterministic and impossible for us to predict.

That's what you said. Not difficult. Not hard. Not beyond our current technology. Impossible. You are making a statement, a factual statement, about impossible things.

That's why I'm calling it semantic masturbation. You're not doing anything useful to explain anything. Stochastic is a fine word. Chaotic works perfectly well. Deterministic but impossible to actually determine is a bit wordy, but sure.

Stating that I'm right, but you don't like the language I'm using while I'm right is just semantics.

[my_wan]

Originally Posted by GreyICE

No, it's of no theoretical value. Brownian motion is an observed phenomena. Explaining it is obviously of value. You are taking an observed phenomena and stating, as an article of faith that it is both deterministic and impossible for us to predict.

That's what you said. Not difficult. Not hard. Not beyond our current technology. Impossible. You are making a statement, a factual statement, about impossible things.

No, you are distorting what I said. I said we can't know it is true for every possible mechanism in the Universe. Not impossible, nor potentially without empirical meaning, for any given model. Though in principle it could be true for some particular model. When a deterministic model adds to the empirical content, which it can in some cases, we predict things like Brownian motion. It is therefore not an article of faith, it is an important theoretical consideration, sometimes with important empirical consequences.

Originally Posted by GreyICE

That's why I'm calling it semantic masturbation. You're not doing anything useful to explain anything. Stochastic is a fine word. Chaotic works perfectly well. Deterministic but impossible to actually determine is a bit wordy, but sure.

Determinism was the guiding principle in the development of statistical mechanics, and this defines its differences wrt classical thermodynamics that didn't take microscopic mechanics into account. Without it there would be no quantitative explanatory model for Brownian motion. We could qualitatively claim it was microscopic collisions, but the disparate scales would make that more than a little suspect. Again, I did not say it was impossible to determine for any given case, I said it's impossible to determine for every possible case. In other words it's impossible to say everything about the entire Universe is deterministic

Originally Posted by GreyICE

Stating that I'm right, but you don't like the language I'm using while I'm right is just semantics.

How right you are wrt to any given model is limited to the accuracy of that model. By rejecting the formal limitations of the model, and assuming randomness means non-deterministic, you are falsely rejecting that these formal limitations exist. What are these formal limitations? They are formalized in the notion of an ensemble.

Originally Posted by http://en.wikipedia.org/wiki/Statistical_ensemble

...is an idealization consisting of a large number of mental copies (sometimes infinitely many) of a system, considered all at once, each of which represents a possible state that the real system might be in. This article treats the notion of ensembles in a mathematically rigorous fashion, although relevant physical aspects will be mentioned.

So what are these "physical aspects" to be mentioned?

Originally Posted by http://en.wikipedia.org/wiki/Statistical_ensemble

Physical considerations
The ensemble formalises the notion that a physicist repeating an experiment again and again under the same macroscopic conditions, but unable to control the microscopic details, may expect to observe a range of different outcomes.

So an ensemble is a formalized method of leaving microscopic states undefined. It must remain undefined in order to maintain generality. Yet in any particular model these microscopic states presumably exist, and define the way the statistics are expressed. Undefined does not mean non-existent causal mechanisms. Yet we cannot prove that in every possible case there exist microscopic states that causally defines the statistics. We only know that in all the cases we have proved there exist causal mechanisms that define these states.

An equally serious problem is that, by rejecting these formal limitations, people in these debates are constantly claiming people made claims they didn't make, calling each other liars, et., etc. Due simply to some people are talking about mechanisms that the formal ensemble leaves undefined, as it relates to a given model, while others are rejecting the notion that anything is left undefined. Defining the unique causal mechanisms of a particular model can have empirical consequences that extend the accuracy of the model, as it did with Brownian motion.

So how right you are is contingent upon the accuracy of the model and the physical limitations of measurement not improving to a point of exposing possible inaccuracies. Those who accept these limitions as undefined, rather than non-existent, will be the ones that improve the models. Even if in some case randomness really does have non-existent causal mechansims, who want to limit themselves from discovering and improving these models simply because of an exclusive definition we chose.

[GreyICE]

Originally Posted by my_wan

No, you are distorting what I said. I said we can't know it is true for every possible mechanism in the Universe. Not impossible, nor potentially without empirical meaning, for any given model. Though in principle it could be true for some particular model. When a deterministic model adds to the empirical content, which it can in some cases, we predict things like Brownian motion. It is therefore not an article of faith, it is an important theoretical consideration, sometimes with important empirical consequences.

We're not talking about the universe, or Brownian motion. We're talking about Evolution. Which does not act deterministically.

The fact that we observe that under certain conditions (like, presence of life, generations, timeframes, pressures, etc.) evolution will occur, and can make generalizations about its nature does not make it deterministic. Pretending evolution is something as pretty and simple as "Insert tab A into slot B, wait 52,768 generations, observe mutation X" is silly.

Quote:

Determinism was the guiding principle in the development of statistical mechanics, and this defines its differences wrt classical thermodynamics that didn't take microscopic mechanics into account. Without it there would be no quantitative explanatory model for Brownian motion. We could qualitatively claim it was microscopic collisions, but the disparate scales would make that more than a little suspect. Again, I did not say it was impossible to determine for any given case, I said it's impossible to determine for every possible case. In other words it's impossible to say everything about the entire Universe is deterministic

No, no. You said it was impossible to say everything about our petri dishes could even be termed identical.

I've heard of relativism, but reducing the entire universe to a couple petri dishes seems a bit much.

Quote:

How right you are wrt to any given model is limited to the accuracy of that model. By rejecting the formal limitations of the model, and assuming randomness means non-deterministic, you are falsely rejecting that these formal limitations exist. What are these formal limitations? They are formalized in the notion of an ensemble.

Nonsense. For non-chaotic systems, the accuracy of the inputs determines the accuracy of your knowledge of the outputs.

Chaotic systems are not so nice. We can generalize them as stochastic, deterministic but impossible to determine, what have you.

The final point remains. You seem to have latched onto a semantic point, and are insisting that evolution is deterministic but impossible to determine.

If you have any objections with the term "deterministic but impossible to determine" feel free to voice them. Otherwise I will be sticking with the more descriptive and less verbose stochastic.

P.S. Random is right out.

[my_wan]

Originally Posted by GreyICE

We're not talking about the universe, or Brownian motion. We're talking about Evolution. Which does not act deterministically.

http://biocomplexity.indiana.edu/jgl...lecture4v2.ppt

Originally Posted by http://biocomplexity.indiana.edu/jglazier/courses/p548/files/p548_lecture4v2.ppt

Monte Carlo Methods Use Statistical Physics Techniques to Solve Problems that are Difficult or Inconvenient to Solve Deterministically.
...
Both Applications Important in Biology.

It appears that you ignored the undefined for a reason part of the formalization of ensembles in my last post.

Originally Posted by GreyICE

The fact that we observe that under certain conditions (like, presence of life, generations, timeframes, pressures, etc.) evolution will occur, and can make generalizations about its nature does not make it deterministic. Pretending evolution is something as pretty and simple as "Insert tab A into slot B, wait 52,768 generations, observe mutation X" is silly.

Well duh... I made no such pretension. However, can you deny that by increasing the level of control of the micro-states, even to a tiny degree, can impart some control on variables that otherwise would have been purely random? We have even developed methods of controlling radioactive decay rates to some degree, but that's not NS is it.

Originally Posted by GreyICE

No, no. You said it was impossible to say everything about our petri dishes could even be termed identical.

Oh.. Well duh... You generalized this to say I claimed it was "deterministic but impossible to actually determine"? That's a freaking stretch. Do you really claim that those petri dishes are perfectly identical? Do you really claim that control can't be increased to some small degree that has empirical effects, over and above previous runs?

Originally Posted by GreyICE

I've heard of relativism, but reducing the entire universe to a couple petri dishes seems a bit much.

Your control in producing identical dishes is woefully inadequate to call them identical. It can pretty much be guaranteed that whatever control you use, it can be improved for dishes even more alike, yet never identical. Even tiny improvements can have empirical effects.

Originally Posted by GreyICE

Nonsense. For non-chaotic systems, the accuracy of the inputs determines the accuracy of your knowledge of the outputs.

Yet even chaotic systems can be deterministic.
http://ptp.ipap.jp/link?PTP/64/1532/

Originally Posted by http://ptp.ipap.jp/link?PTP/64/1532/

Statistical Mechanics of Deterministic Chaos

Originally Posted by GreyICE

Chaotic systems are not so nice. We can generalize them as stochastic, deterministic but impossible to determine, what have you.

Yes, and even small improvements in measurement has allowed people to beat roulette wheels.
http://www.iht.com/articles/1992/05/05/chao.php

Originally Posted by GreyICE

The final point remains. You seem to have latched onto a semantic point, and are insisting that evolution is deterministic but impossible to determine.

No. I explained in detail why it is not totally semantics. Claiming this is a semantics argument to keep falsely accusing me and other of claims we did not make. Also explained in detail was this false notion that it is always impossible to determine. Neither am I insisting that evolution or anything else MUST be deterministic.

I suggest you read and respond to those explanations, rather than repeating the same false BS. If you make such naked claims about what I claimed again, I'll call your claim a bald faced lie.

Originally Posted by GreyICE

If you have any objections with the term "deterministic but impossible to determine" feel free to voice them. Otherwise I will be sticking with the more descriptive and less verbose stochastic.

Duhhh.. Let's see, what did I say?

Originally Posted by my_wan

...Defining the unique causal mechanisms of a particular model can have empirical consequences that extend the accuracy of the model, as it did with Brownian motion.

So how right you are is contingent upon the accuracy of the model and the physical limitations of measurement not improving to a point of exposing possible inaccuracies. Those who accept these limitations as undefined, rather than non-existent, will be the ones that improve the models. Even if in some case randomness really does have non-existent causal mechanisms, who want to limit themselves from discovering and improving these models simply because of an exclusive definition we chose.

Where in there was "impossible to determine? And this was my particular post you responded to with that crap. Nor did I ever claim that everything is fully deterministic. I don't want to hear this claim again.

Originally Posted by GreyICE

P.S. Random is right out.

Given the biocomplexity.indiana.edu reference above the Boltzmann Equation is in fact relevant, as this equation defines statistical mechanics, irrespective of field of application. This reference should be of interest then:

Originally Posted by http://mathema.tician.de/news.tiker.net/download/academic/talks/boltzmann-slides.pdf

Deterministic Methods for the Boltzmann Equation

Are these people just masturbating themselves?

Edit: Clickable link for that last quote
http://mathema.tician.de/news.tiker....ann-slides.pdf

[cyborg]

Originally Posted by jimbob

So the course of evolution is sometimes altered by random mutations, which are inherent to the system.

That's an entirely different statement.

Quote:

The fitness landscape is subject to random change. Which I now think you agree with.

That is also an entirely different statement.

Quote:

So are you saying that for a given stable ecosystem, that evolution is nonrandom?

No.

What I'm saying is pretty simple. I really don't understand how you can infer such tangential meanings from it.

[jimbob]

What are you saying?

Originally Posted by jimbob

Well I read it as this:

Originally Posted by cyborg

Originally Posted by jimbob

So we have accepted that mutations can cause significant differences in outcomes.

Uh, I've never said anything different.

So the course of evolution is sometimes altered by random mutations, which are inherent to the system.

Originally Posted by cyborg

Originally Posted by jimbob

So the course of evolution is sometimes altered by random mutations, which are inherent to the system.

That's an entirely different statement.

What are you saying?

So we have accepted that mutations can cause significant differences in outcomes. But the course of evolution isn't affected by this despite the fact that it causes different outcomes?

Originally Posted by cyborg

Quote:

The fitness landscape is subject to random change. Which I now think you agree with.

That is also an entirely different statement.

OK, so you don't agree with that? I would say that the evolution of citrate⁺ metabolism changed the fitness landscape for other organisms within the ecosystem, why do you disagree?

Originally Posted by cyborg

Quote:

So are you saying that for a given stable ecosystem, that evolution is nonrandom?

No.

What I'm saying is pretty simple. I really don't understand how you can infer such tangential meanings from it.

What you are saying *might* be pretty simple, however you prefer to allude to it, rather than state it outright.

Hey, I'd say that for a given stabe ecosystem, over moderate timescales, with sufficiently large populations, then randomness is often unimportant in evolution, even if it is still implicitly present.

[cyborg]

Quote:

What are you saying?

So we have accepted that mutations can cause significant differences in outcomes. But the course of evolution isn't affected by this despite the fact that it causes different outcomes?

Again: I have not "accepted" this because I've never said anything to the contrary. I would like to know why you think otherwise. It is almost certainly because you've never understood what I've been saying and have been confusing the application of certain concepts to levels of abstraction I was not applying them to.

Quote:

OK, so you don't agree with that? I would say that the evolution of citrate+ metabolism changed the fitness landscape for other organisms within the ecosystem, why do you disagree?

Since I don't disagree why do you think I do?

Quote:

What you are saying *might* be pretty simple, however you prefer to allude to it, rather than state it outright.

I've stated it outright hundreds of times.

Quote:

Hey, I'd say that for a given stabe ecosystem, over moderate timescales, with sufficiently large populations, then randomness is often unimportant in evolution, even if it is still implicitly present.

You do not understand what I've been saying. The overall behaviour of the system doesn't imply a particular behaviour of its constituents. I've proven this before. Either you don't want to understand or can't.

[jimbob]

Originally Posted by my_wan

For this debate, or peoples characterizations of others claims, to make any sense whatsoever depends on what definitions are used for what.

cyborg, jimbob, etc., answer these questions.

(1) Is random and deterministic antonyms?
(2) Does randomness rule out determinism?
(3) Is your answers here model specific or do they also apply to the actual system being modeled?

Differences in how you answer these questions determines what others think you said that you didn't. This leads to accusations of false claims, lying, etc., etc. So answer them or waste your time.

My_wan:

(1) Is random and deterministic antonyms?
Yes: In a deterministic system, identical inputs produce identical outputs. *similar* inputs can produce grossly dissimilar outputs, and this can still be determistic.

In a random system, identical inputs needn't produce identical outputs. For example, it really does seem as if radioactive decay is random. Furthermore, some of the double-slit experiments, it seems as if the universe itself hasn't decided on the outcome of a quantum event before it happens. In other words, these events are truely random, and not just unpredictible.


(2) Does randomness rule out determinism?
It depends on the system, but in evolution, I'd say that the course of evolution can be afffected by (say) a "disruptive mutation" that confers such selective advantage, that it alters the entire fitness landscape. This exists potentially within any evolutionary system, even though it wouldm't occur often: To use Lenski's own words:

Quote:

In any case, our study shows that historical contingency can have a profound and lasting impact under the simplest, and thus most stringent, conditions in which initially identical populations evolve in identical environments. Even from so simple a beginning, small happenstances of history may lead populations along different evolutionary paths. A potentiated cell took the one less traveled by, and that has made all the difference.

" small happenstances of history may lead populations along different evolutionary paths. A potentiated cell took the one less traveled by"

In this case, the "small happenstances of history" were simply differences in mutations, as the rest had been controlled for in the experimantal design.


(3) Is your answers here model specific or do they also apply to the actual system being modeled?
I am talking about evolution as it has been observed. I would argue that it is probably inherent in any highly nonlinear system if miroscopic random events can get magnified (in a determinsitic fashion).

I would argue that if you have self-replicating systems with errors in copying, then Darwinian evolution will follow, and that if these errors in copying are random, then there is the potential for a disruptive "mutation", so any system with self-replication with random errors will have the potential for random (grossly different) outcomes.

[cyborg]

Originally Posted by my_wan

(1) Is random and deterministic antonyms?

Yes.

Quote:

(2) Does randomness rule out determinism?

No.

Quote:

(3) Is your answers here model specific or do they also apply to the actual system being modeled?

Indeterminable.

[GreyICE]

My_wang, can you possibly add something useful? You seem to be attacking me along the concept that evolution MUST be deterministic (I can't really tell, you've butchered my arguments long past the point of readability). But... why? Given what can more or less be described as identical inputs, we have predictable outputs, but the predictability and is marred by what can be described as a chaotic, or a random factor.

This is perfectly described by the term stochastic, which is a process that is model-able, but contains a random element that has to be accounted for.

You have posted lots and lots of links, lots and lots of wasted verbage, lots and lots of nonsense about how improvements to the modeling of the 'chaotic' factor improves our ability to predict the outcome of the system.

Why yes, yes it does. That does not make the behavior deterministic. This does not allow a set of error bars on the inputs to determine the set of error bars on the outputs.

Also, you are unfairly setting up random and deterministic as the ONLY possibilities. Systems can be neither. Your chaotic modeled as deterministic simply seems to be saying that they can model the outcome of some chaotic systems. Your second article was about a group of bloody investment bankers. Guess what, 10:1 they're out of business right now. How'd predicting the chaotic work out for them?

My final question to you: Is there any objection to using the term stochastic to describe evolution that CANNOT be summarized using the word "semantic?"

[my_wan]

jimbob,
That's a fair position. Here are the reasons you mistakenly misjudge cyborg's statements. It is not provable one way or the other if, in some cases, whether randomness is an antonym of determinism or not. We know certain processes like, ideal gas laws, can be formulated as perfectly random systems. So in that case randomness and determinism is not antonyms. Yet we don't know that this can be extended to all systems. Like the ones you mentioned. Randomness in those cases may not in fact be an antonym of determinism. Yet we must know the things that irreducible randomness says doesn't exist to prove that it doesn't exist, in those cases we just don't know.

This creates a "defensible" position that all randomness is the result underlying deterministic causes. Yet it's not any more provable, in all cases, than the contrary. So when cyborg says randomness says nothing about the causality of the mutations themselves, his position makes sense. It neither accepts nor denies that randomness and determinism are antonyms. By assuming they are definitely antonyms it gives meaning to cyborg's statements that cyborg was not intending to imply.

I don't really care what definition you choose, so long as people learn how to parse the contrary definitions and quit falsely claiming someone else claimed something they didn't. This of course goes to both sides of the debate.

[my_wan]

GreyICE,
Yes I was a bit harsh. The point is that the epistemic differences in how people choose to define randomness. These differences hinge on epistemic notions of how it relates to determinism. It is important in a debate to at least know the opponents position even if you disagree. Here are the things that can, honestly and without taking sides, be said about randomness.

(1) Some perfectly determistic models can also be modeled as purely random.
(2) When a deterministic model is not known, we can't prove one doesn't exist.
(3) Determistic causes sometimes has empirical differences wrt certain variables.
(4) Determistic causes sometimes has no empirical differences wrt certain variables.
(5) Determistic causes sometimes "might" not exist at all wrt certain variables.

That all we can say given the fundamental limitations of both math and science.

By locking onto a singular definition, and defending it as the only definition, it forces false characterizations of the statements of others useing a contrary definition. Either of the definition may or may not be true with respect to a given phenomena. It's the old correlation does not equal causation, only the correlation the lack of correlation this time. Neither does this rule say correlation does not equal causation in any singular case. I drop a glass on the floor and it breaks, then correlation does equal causation in that case.

The ensuing debate is nothing more than semantics until this is understood. Once this is understood debate is possible.

[GreyICE]

Originally Posted by my_wan

GreyICE,
Yes I was a bit harsh. The point is that the epistemic differences in how people choose to define randomness. These differences hinge on epistemic notions of how it relates to determinism. It is important in a debate to at least know the opponents position even if you disagree. Here are the things that can, honestly and without taking sides, be said about randomness.

(1) Some perfectly determistic models can also be modeled as purely random.
(2) When a deterministic model is not known, we can't prove one doesn't exist.
(3) Determistic causes sometimes has empirical differences wrt certain variables.
(4) Determistic causes sometimes has no empirical differences wrt certain variables.
(5) Determistic causes sometimes "might" not exist at all wrt certain variables.

That all we can say given the fundamental limitations of both math and science.

By locking onto a singular definition, and defending it as the only definition, it forces false characterizations of the statements of others useing a contrary definition. Either of the definition may or may not be true with respect to a given phenomena. It's the old correlation does not equal causation, only the correlation the lack of correlation this time. Neither does this rule say correlation does not equal causation in any singular case. I drop a glass on the floor and it breaks, then correlation does equal causation in that case.

The ensuing debate is nothing more than semantics until this is understood. Once this is understood debate is possible.

1) deterministic systems are ones in which no randomness is part of the future outcome of the system. Period. Models are models, they frequently contain simplifications.
2) True, and irrelevant. If one cannot perfectly predict the outcome and one cannot replicate the outcome reliably, it is fair and fine to call the system non-deterministic. If a method later crops up that allows you to call the system deterministic, that's well and good, but stating that such a method exists on zero evidence is not good reasoning.
3) Obviously. What they do not have is a random element. One does not pretend that all velocities result in equal distance traveled for equal time, but one does not pretend that distance as a function of velocity contains a random element. You are stating the obvious.
4) True. Also irrelevant.
5) Okay, so not all systems are deterministic?

Once again, there is no evidence that evolution is a deterministic system, and plenty of evidence that there are uncontrollable elements that resemble random. Stochastic is a fine term, and one I rather like. Random is a stupid term, totally wrong, and used by idiots. Chaotic works in my book, but I like chaotic systems overly much, they appeal to me.

[recursive prophet]

Interesting debates the last couple days, and I’d like to thank all the participants for contributing to my education. One thing I still wonder about is if there is anyone here that would argue-wrt the OP-that natural selection isn’t stochastic.

I must confess this extended discussion is getting quite difficult for my aging mind to comprehend. As to my_wan’s 3 questions, I would answer yes, no, and undetermined. My answers might also apply to real systems, but Cyborg may be correct in saying they are indeterminable, and I find Jimbobs answer quite compelling.

I have posted many of the comments from this thread on a similar one at RDF, and then their replies here. However, the volume of recent activity on RDF would sort of swamp the discussion here, so this time I am going to just post links on both sites. I would strongly encourage everyone to read what has been written there, especially the most recent exchanges between my_wan and susu.exp. The latter is working toward his doctorate in paleontology at Bonn University, and it appears has some views that are congruent with those of GreyICE. He also responded to Sol’s last post here, and I would really like to read both of their reactions to what he and others there have to say.

To read all my_wan wrote when he recently reentered the debate on RDF, click here, or to jump ahead to where susu comes in with his rebuttals, use this link. I look forward to reading more of this hybrid-enhanced dialectic, but will withhold any commentary for now as after digesting my_wan’s latest posts on RDF I’m thoroughly confused. I'm well accustomed to being the dumbest guy in the forums I choose to read, but this one is making me consider early onset dementia.

[articulett]

Evolution is deterministic... all science is on a macro scale-- we know that what unfolds is caused by what came before. When we investigate an air disaster it's because we understand that figuring out the cause, helps us prevent such things in the future. We know that the traits we see evolved because ancestors that had such traits preferentially survived and reproduced over their peers that didn't.

Understanding evolution involved understanding the way what we see unfolded-- In this sense, it is deterministic-- what happens has a cause which has a cause which has a cause...

The reason my_wan and others are forced to delineate this is because --to Mijo, RP, and Jim-Bob, stochastic = random (because stochastic processes are sometimes called "random processes"). And anything that has any randomness in it, is wholly random to them. They are not using the same definitions as others. You don't need to use the words random, deterministic, nor stochastic to describe evolution-- for evolution to occur you just need variety and an environment that selects some information to be copied over others-- that's it. My_wan has gone out of his way to be more than clear on this topic. The problem is that people have varying definitions of random such that every process can be said to have some randomness according to some definitions of random. This makes all processes "random" and nothing deterministic. It renders the terms useless for generating understanding. That is why people who cannot answer Sol's simple query the way others would should cannot participate in the same conversation. Words mean different things to them.

My_wan's not the one playing around with semantics on this topic. He's clarifying misstatements that have been attributed to him. My_wan is explaining why it's misleading and uninformative to call evolution random--he's not saying it IS deterministic-- though, at least on a macro level--everything is deterministic in science-- that is, everything is determined by what came before-- we understand the past by putting the pieces of the puzzle together, because we know that things did not happen "magically" or without physical causes. Evolution is not a random process-- it is directed by what information gets copied the best in whatever environment it finds itself in. The genes that get passed on the best DETERMINE the direction of what evolves. It's no more random than the winner of a chess tournament or an election. Calling it random is like saying the winner of a chess tournament is chosen at random. It's misleading, vague, and daft.

Allele selection is stochastic to some extent, but not always --as the paper in Current Biology illustrates http://www.bio-medicine.org/biology-...-study-1268-1/

It really depends on how you are defining the terms and in respect to what. But evolution is mainly driven by genes and genomes that confer a survival/reproductive advantage to their owners. Stochastic really doesn't capture that understanding. It's useful for understanding drift, but drift appears to play a much smaller role than once imagined. Eyes did not evolve by drift, for example. Nor did Lenski's bacteria. Each step in the mutation provided a slight reproductive advantage to the population that had it... My_wan has a world of science supporting his statements. There are no peer reviewed scientists saying "evolution is random (full stop)" or even "evolution is stochastic" (period)-- only that some aspects can be modeled scholastically. But so can election outcomes and tons of stuff. That doesn't mean it is informative to say "elections are stochastic". The same with evolution. It's even more meaningless to say evolution is stochastic than to say elections are stochastic. It transmits no information as to what evolution actually entails, and it's misleading. It makes people feel like they are saying something intelligent--but to the educated ear, it sounds like they think they understand evolution much better than they actually do!


It's preferential survival and reproduction that is the method by which genomes evolve and thus the organisms they code for. But each organism dies the same species it is born as. It's genome has a chance to evolve based on how successful the organism is in passing on it's genes. I don't think "stochastic" captures this understanding. That is why few of the experts in the field use it to explain evolution. And none call evolution random. Perhaps few use stochastic because it makes some not-so-bright people think that they are calling evolution "random"--it obfuscates rather than clarifies. It's not "wrong", but it's about as informative as calling evolution a "noun". It's not that stochastic can't be used to explain evolution. It's just that some people are defining words so loosely and bizarrely that saying such is completely uninformative regarding the most essential aspects of understanding evolution. If you want to describe evolution to someone, I suggest you use the words of experts like Ken Miller, Richard Dawkins, Charles Darwin, etc. They are responsible to conveying understanding to millions. I don't think anyone calling evolution "random" or "stochastic" can claim such. In fact, they are describing evolution similarly to those who are well known for obfuscating understanding in order to support "intelligent design". To those who understand the topic, they sound like they haven't got a clue as the quotes from Miller, Dawkins, Ayala, Coyne, et. al. reiterate.

Many scientists feel that nothing is truly random-- but they aren't defining random as "anything to do with probabilities" the way some vague folks are. They have a much more specific definition in mind as My_wan has repeatedly tried to clarify to no avail. Mutations aren't random in that they have causes and some areas of a genome are far more prone to mutation (hot spots) than others-- and some types of mutations are far more likely than others-- but genomes have been "selected" to be this way. They are just "random"-- or rather unbiased in that they happen without regard to whether they'll benefit an organism or not. And then the environment selects. But we know why or how many changes happen even if we can't predict where or when. We understand things like non-disjunction. They have physical causes--in this way, they are NOT random. We can predict that older mothers will conceive more trisomies for example-- this isn't "random"-- but we can't predict much more specifically than that. But because we live in a deterministic world, we can discover and predict better and better and understand more and more.

[my_wan]

Originally Posted by GreyICE

1) deterministic systems are ones in which no randomness is part of the future outcome of the system. Period. Models are models, they frequently contain simplifications.

Here you took a definite position and punctuated with an unequivocation. Then the second sentence you equivocated, except you said nothing about how this equivocation might effect the first sentence.

Originally Posted by GreyICE

3) Obviously. What they do not have is a random element. One does not pretend that all velocities result in equal distance traveled for equal time, but one does not pretend that distance as a function of velocity contains a random element. You are stating the obvious.

Apparently I didn't make this clear enough. All 5 of those statements were specific to systems that could be modeled as random.

Originally Posted by GreyICE

5) Okay, so not all systems are deterministic?

This one just makes me cringe. The point is You don't know, I don't know, nobody knows.

Consider a machine with a red and blue light, and a button. You press the button and either the red or blue light comes on. Press again and it goes off. Let's say you kept notes for which light came on every time for thousands of hours. You get a perfectly random 50/50 normal distribution of red and blue lights. Does this mean the machine randomly selects which light to turn on? Does that mean it is unpredictable and there is no way to get it to give you the same color light every time, without changing anything about what or how the machine works? No, it doesn't! With enough precision and control you can get it to turn the same light on every time. It simply switches the between red and blue every 1/1000th of a second. So once you know this and have accurate enough a clock and button presser the color you get is predetermined even 10 years from now.

So in you response to 1), how do you know when the randomness indicates the future is indeterminable and when you model simply lacks the details to predict it? The way you stated it makes this even worse. You said: "determinism means 'no' randomness is part of the future outcome". Yet the red and blue light was perfectly random yet perfectly deterministic. Lots of different deterministic phenomena exhibit this same type of randomness, ideal gas laws for instance. Of course a real gas is not so perfect as the ideal model but the randomness is the same. So we can not know if all randomness is the result of underlying deterministic parts. It would require knowledge as perfect as that some presume sky daddy possesses.

The most important thing to learn is if someone claims NS is deterministic, they did not necessarily claim it is not random, just like the red and blue lights. Insisting they did is wrong. Accusing others of claims they did not make is the primary effect. It's a waste of time, and a diversion, to insist on debating "is NS random" when the argument is actually about the definition of randomness. So what exactly is the differences in the definitions?

Always Deterministic Definition
Definition(1): Randomness and determinism is synonymous. Randomness is a formalized methodology of leaving data, generally microscopic, undefined while maintaining data about the averages and distribution of that microscopic data. This entails, by the epistemology of some, that all randomness is pseudorandom, even when, due to undefined data, it is mathematically perfectly random.

Sometimes Deterministic, Sometimes Random Definition
Definition(2): Randomness and determinism are antonyms. Yet deterministic systems can exhibit pseudorandomness due to undefined data. Depending on ones epistemology it may or may not entail that mathematically perfectly randomness can also be pseudorandom.

Always Random Definition
Definition(3): Randomness and determinism are antonyms. Anything that exhibits randomness is in fact random, not pseudorandom. Anything that exhibits pseudorandomness has a defined deterministic causal mechanism. However complex and computationally hard that mechanism is, pseudorandomness allows this mechanism to be recovered from deviations from perfect randomness. In effect the random and nonrandom elements can be separated.

Problems with Definition(1): This is philosophically defensible, yet impossible to prove. It requires perfect knowledge of the entire Universe to prove.

Problems with Definition(2): If the epistemology states that mathematically perfect randomness can be pseudorandom then this definition entails that we can't, even in principle, alway know the difference between randomness and pseudorandomness, though sometimes we can. So functionally it similar to Definition(1). If the epistemology is the converse, then our capacity to know the difference between randomness and pseudorandomness is contingent upon the sample size (largeness of N). So we can only place increasing odds on whether it is truly random, but when it is found to be pseudorandom it is verifiably so.

Problems with Definition(3): This definition entails that all systems are definitely either random or pseudorandom, and randomness entails purely random. I am less sure of the epistemological variances of the advocates. This is made more difficult when an unequivical statement like: "deterministic systems are ones in which no randomness is part of the future outcome of the system. Period.", is followed by 'potential' equivocations like: "Models are models, they frequently contain simplifications".

What can we empirically say? We can construct a perfectly idealized deterministic models. It doesn't matter whether it represents a real model of physical laws or not, it's construction entails that it is real in some sense. The idealization simply entails that we can't generalize it to all models. Yet, when we intentional leave certain parameters of this perfectly deterministic system undefined, it can exhibit pure randomness. Ideal Gas Laws are an example of this, and Statistical Mechanics is the result when we leave microscopic detail undefined. At the microscopic level it's just a 3-D billiards game.

So we can empirically say determinism can exhibit perfect randomness, if the information is limited. Yet because these models are only idealizations, we know both the epistemology and Definition(3) wrt very limited singular cases is wrong, but may in fact be correct for some systems. To prove it requires the very information that Definition(3) says doesn't exist. Empirically we are therefore left somewhere between Definition(1) and (2). Although Definition(3) may apply in a subset of systems and/or variables.

I find it easier to presume Definition(1) for the same reason I presume naturalism. This is not a truth statement though, because I am limited by methodology the same way naturalism is limited to methodological naturalism. If in particular case I am wrong, it can never be proved. In those cases where I am right it can extend the accuracy of the models and/or expand our empirical knowledge, the same way Einstein did with Brownian motion.

[jimbob]

Originally Posted by articulett

Historical contingency may mean "random" to you-- it doesn't mean it to Lenski or anyone else. Historical contingency IS the selection process--and nobody of note would describe that as random.

Articulett, in the Long Term Evolution Experiment, the experimental design controlled for the other factors, so that random mutations were shown to be the important difference between the populations that evolved the trait, and those that didn't.

Only in one flask did a "neutral" potentiating mutation arise that later on facilitated the evolution of citrate⁺ metabolism.

In this experiment, the important "historical contingencies" were simply the differences in the mutations in the different populations. These were random. This means that the occurance of a random mutation allowed later evolution of a trait.

It happened in a very simple ecosystem, and would be more common in more complex ecosystems with more compex interactions.


I agree with the New Scientist article about this:

Quote:

In the meantime, the experiment stands as proof that evolution does not always lead to the best possible outcome. Instead, a chance event can sometimes open evolutionary doors for one population that remain forever closed to other populations with different histories.

This is practically what I had written here (note the dates..)

Quote:

Selection is probabilistic, but this still leads to optimisation to environments (although the nature of the optiimisations will depend on mutations, and might be surprising, or indeed diferent if it were posible to rreun the situation, to identical conditions, as the quantum events would still be different).

That optimisation to environmental stimuli will occur is inevitible, what optimisation isn't.

Over geological timescales, I would argue that the environment itself is random. Some orbits are chaotic, should that include a potential "dionosaur killer" asteroid, then quantum events could influence asteroid strikes on earth. As organisms fill empty niches, they prevent other organisms filling slightly differnet niches; whilst lions exist, no leopard descendant will evolve into something that competes completely with lions.

[articulett]

Only to you Jim-bob. I prefer the words of actual experts to your verbiage. I'll take Miller, Dawkins, et. al. over your self appointed expertise any day.

To me, you sound like you think you understand evolution and Lenski much better than you actually do. To you, Lenski's paper is proving that evolution is "random". I don't think any actual expert would think that Lenski's paper proved that evolution is "random" any more than it "proves" evolution is a noun. You are just so bizarrely hyperfocused on the randomness that you miss the essence of what others are saying again and again and again. Like my sig article, you have no ability to recognize your own general competence in the subject and so you miss the opportunities to ameliorate your ignorance. You think you know more than those who might clue you in.

And you haven't progressed an iota in 2 years when it comes to ability to understand or engage in conversation on this topic in my opinion. Cyborg spoon feeds you the information, and you just cannot digest it. I think you need to stick with conversing with those who find you as much of an "expert" as you imagine yourself. I don't think you are any more capable of explaining evolution to anyone than Behe is, though I'm sure your confidence in your ability to do so is on par with his.

You just seem to be in your own world on this topic-- or rather in the same world as Behe, T'ai Chi, Mijo, RP, etc. I think that makes you suitable for conversation with each other, but not particularly good at conveying information to others or learning from those who understand and communicate better than you. Your goal is to prove that your way of explaining things is useful or correct or right or informative... but it's just not. You sound identical to folks like Dembski. Who can converse with someone like that on this topic? And why would any serious scientist or person who actually understood evolution do so? Why can't you perceive how much you sound like them and how little you sound like those who actually teach others and write extensively on this subject? How can you imagine you are more right or clearer than Ken Miller, for example??

[my_wan]

jimbob,
When you read a research article that uses terms like "chance event", "random", etc., you need to realize that the use of the term is not implying any singular epistemological notions of what those terms mean. They have a purely operational meaning, especially in biological sciences. Operationally the difference between what's included in the model and what the actual state of affairs are in the dish is nil, unless you can find a way to include perfect, or at least more concise, information in the model.

The mistake you make is adding your own epistemology to these terms, pretending that these researchers are using the same epistemology in the use of those terms, then accuse others of statements they didn't make because they didn't share the same epistemology. The term "chance event" was epistemologically neutral as used by that article. Operationally it simply means: an event in which it is "unknown" what the cause was, or if it even had a cause. It simply means "we don't know".

So quit pretending you know, and can tell others they are wrong, and that they said things they didn't say.

[jimbob]

Articulett,

One of Lenski's stated aims was to see whether rerunning the tape of evolution would always produce the same results every time. It won't.

I guess you disagree with the following statement:

In an evolutionary system, where the only differences are due to random events, significantly differenct outcomes can be observed. Indeed this has been observed in the Long Term Evolution Experiment, where significant differences in outcomes have been observed, and where minor differences in the initial conditions have been shown to be unimportant to this outcome.

If you do disagree with this, could you explain why only one population evolved citrate⁺ metabolism? What were the important differences in this population that were not due to differences in mutations?

ETA:

Do you disagree with this:

However, selection requires heritable variation generated by random mutation, and even beneficial mutations may be lost by random drift. Moreover, random and deterministic
processes become intertwined over time such that future alternatives may be contingent on the prior history of an evolving population. For example, multiple beneficial mutations will arise
in some unpredictable order, and those that are substituted first may differ from others in their effects, thus constraining some evolutionary paths while potentiating other outcomes.

[articulett]

Jim-Bob, I don't think you are capable of having a discussion on this topic because you are bent on proving to yourself that you are saying something "intelligent". Sure, it is "valid" to say evolution is a "noun"-- but this validity doesn't make focusing on the fact "intelligent" or useful. The same with your focus on "random". I consider you as off topic and bizarre as someone obsessed with proving evolution is a "noun". Really. And I don't think it's fixable. You and Mijo are the same as you were 2 years ago--the same as Behe was 2 years ago. I'm a skeptic-- I don't have faith that anything will ever change in this area.

So I am going to put you on ignore once more so that you might engage those who are using terminology the way you are and feel that it's useful or informative to describe evolution in the way you do. Remember, to me, I'm about as interested in your obsession as I would be with someone who wants to engage me in a discussion to "prove" that evolution is a noun. To you, evolution is "random". No expert would describe it as such... your twisting their words to hear them saying is much makes you look as dishonest as Behe from my perspective. If you don't want to be treated like an "intelligent design cproponentist", then you really ought to distinguish your conversation techniques from theirs.

I mean I could use your "debate techniques" to prove that Lenski's paper is proof that evolution is a "noun" and it would be on par with what you are doing. It wouldn't be "wrong"-- just obfuscating, daft, and silly. Would you engage in a discussion with a person bent on driving home the point that Lenski's paper proves that evolution is a noun?

I wouldn't either. What would be the point? Such people need to be directed towards other people who find such semantic philosphical "debates" stimulating so they don't muck up understanding of others.

[jimbob]

Originally Posted by my_wan

jimbob,
That's a fair position. Here are the reasons you mistakenly misjudge cyborg's statements. It is not provable one way or the other if, in some cases, whether randomness is an antonym of determinism or not. We know certain processes like, ideal gas laws, can be formulated as perfectly random systems. So in that case randomness and determinism is not antonyms. Yet we don't know that this can be extended to all systems. Like the ones you mentioned. Randomness in those cases may not in fact be an antonym of determinism. Yet we must know the things that irreducible randomness says doesn't exist to prove that it doesn't exist, in those cases we just don't know.

This creates a "defensible" position that all randomness is the result underlying deterministic causes. Yet it's not any more provable, in all cases, than the contrary. So when cyborg says randomness says nothing about the causality of the mutations themselves, his position makes sense. It neither accepts nor denies that randomness and determinism are antonyms. By assuming they are definitely antonyms it gives meaning to cyborg's statements that cyborg was not intending to imply.

I don't really care what definition you choose, so long as people learn how to parse the contrary definitions and quit falsely claiming someone else claimed something they didn't. This of course goes to both sides of the debate.

So when cyborg says randomness says nothing about the causality of the mutations themselves, his position makes sense. It neither accepts nor denies that randomness and determinism are antonyms.

Fine, except for:

Originally Posted by cyborg

Originally Posted by my_wan

(1) Is random and deterministic antonyms?

Yes.

Originally Posted by my_wan

jimbob,
When you read a research article that uses terms like "chance event", "random", etc., you need to realize that the use of the term is not implying any singular epistemological notions of what those terms mean. They have a purely operational meaning, especially in biological sciences. Operationally the difference between what's included in the model and what the actual state of affairs are in the dish is nil, unless you can find a way to include perfect, or at least more concise, information in the model.

The mistake you make is adding your own epistemology to these terms, pretending that these researchers are using the same epistemology in the use of those terms, then accuse others of statements they didn't make because they didn't share the same epistemology. The term "chance event" was epistemologically neutral as used by that article. Operationally it simply means: an event in which it is "unknown" what the cause was, or if it even had a cause. It simply means "we don't know".

So quit pretending you know, and can tell others they are wrong, and that they said things they didn't say.

I am familiar with statistics where you often treat multiple nonrandom events as random ones because of ease of analysis, I do that a lot in my day job. In fact you can probably argue that the bell curve is a classic distribution that you get in such a situation, where you may not be dealing with true randomness, but simply unpredictibility.

I'd argue that in the case of evolution it is more than "we don't know".

In the Long term Evolution Experiment, the important difference was traced to potentiating mutations in one population. Furthermore, one of the stated aims of the experiment was to see whether rerunning the tape of evolution always gave the same results. The experimental design accounted for extraneous factors except for mutation. This was one of the experimental aims.

Whether you consider mutations to be truly random or not, must depend on whether you consider quantum events to be truly random. I'd also say that although it isn't completely conclusive, most evidence is that quantum events are truly random.

[jimbob]

norty

[cyborg]

Quote:

Whether you consider mutations to be truly random or not, must depend on whether you consider quantum events to be truly random.

No.

[jimbob]

Why?

[cyborg]

Again:

Quote:

You do not understand what I've been saying. The overall behaviour of the system doesn't imply a particular behaviour of its constituents. I've proven this before. Either you don't want to understand or can't.

[Dr Adequate]

Originally Posted by jimbob

Agreed DrA, just that it is less wrong than saying "nonrandom"

If you have to chose a one-word answer...

But I don't.

[my_wan]

Originally Posted by jimbob

Originally Posted by my_wan

So when cyborg says randomness says nothing about the causality of the mutations themselves, his position makes sense. It neither accepts nor denies that randomness and determinism are antonyms.

Fine, except for:

Originally Posted by cyborg

(1) Is random and deterministic antonyms?
Yes.

Yet he also said:

Originally Posted by cyborg

(2) Does randomness rule out determinism?
No.

I can't speak for him, but my take is that he made a distinction between random and randomness. Random being operationally unknown, and randomness being a numerical property that can apply to either case. Yet you chose the one that was easiest to argue. I philosophically disagree with cyborg over statements like:

Originally Posted by cyborg

Originally Posted by GreyICE

P.S. If you clone 15 bacteria in 15 petri dishes with identical nutrients, light conditions, etc - will they all evolve the same features at exactly the same time?

Yes - deterministic
No - Stochastic

Which do you think it is?

Correct - simplistic.

And have argued with GreyICE at length over it. Yet cyborg also said:

Originally Posted by cyborg

No. I've always said that the underlying cause of mutations is irrelevant.

So he's not ultimately denying cause or no cause. Nor does this impose an epistemology on the use of the term "random" in the general use sense. The answer to 1) seems a stricter mathematical sense, but I don't know. I do know he is also guilty of semantically imposing on your epistemology.

Originally Posted by jimbob

I am familiar with statistics where you often treat multiple nonrandom events as random ones because of ease of analysis, I do that a lot in my day job. In fact you can probably argue that the bell curve is a classic distribution that you get in such a situation, where you may not be dealing with true randomness, but simply unpredictibility.

I'd argue that in the case of evolution it is more than "we don't know".

I don't get how "we don't know" can be more than "we don't know". It appears that you are implying that since we tried and still "we don't know", somehow that means the cause is non-existent.

Originally Posted by jimbob

In the Long term Evolution Experiment, the important difference was traced to potentiating mutations in one population. Furthermore, one of the stated aims of the experiment was to see whether rerunning the tape of evolution always gave the same results. The experimental design accounted for extraneous factors except for mutation. This was one of the experimental aims.

So by rerunning this same experiment are you assuming that both started with the same number of free radicals with the same locations and positions wrt the bacteria in all the dishes. I doubt it but it would take far more detail than even that to say a unique outcome of one dish means more than "we don't know".

Originally Posted by jimbob

Whether you consider mutations to be truly random or not, must depend on whether you consider quantum events to be truly random. I'd also say that although it isn't completely conclusive, most evidence is that quantum events are truly random.

Yes these are interesting subjects to me, and these things definitely fall in the category of "we don't know". We can trivially consider QM deterministic simply by defining the wavefunction as a physically real field. This is however cheating, because it doesn't actually provide the causal mechanisms, it just defines it to be so. It also creates a whole buttload of other theoretical issues.

[recursive prophet]

I was hoping the debate between Jimbob and my_wan would continue, though we seem to be stuck both here and on RDF with "we just don't know" so this is actually a philosophical issue.

There are many working assumptions in math and science that rely on an unproven hypothesis or two, and on the RDF sister thread this has been a key issue. Do most biologists approach evolution as primarily stochastic, or deterministic? As I have also requested there, I would like to read my_wan's answer to the 3 questions he posed here earlier, only with the term stochastic replacing random.

For new viewers, there are several links to the longer discussion at Dawkins.net that I posted on the previous page, and you'll find both JB and MW there also.

[Dr Adequate]

Ye have sown the wind ...

[recursive prophet]

Cute, Doc. And don't think I haven't often considered the possibility we are debating Russel's tea pot here. Perhaps I'll never know.

As you seem to have a philosophical bent, here's a question you might be able to answer for me. I've oft seen it mentioned that random evolution is the creationist argument, and the logic escapes me. If in fact our universe is deterministic, wouldn't this help make the case for a 'determiner?' Conversely, if all is chaotic, then how could one argue an omnipotent design? Intended chaos? Seems much more of a stretch that way then to assume all is preordained. If its all random, for what need have we gods?

Anyhoo, if you deign to enlighten me I really am curious about the reasoning here, and intuitively random wins for me every time. It only yields at a certain point to the realm of the stochastic. All Greek to me.

[cyborg]

Originally Posted by recursive prophet

If in fact our universe is deterministic, wouldn't this help make the case for a 'determiner?'

Since one does not tend to ascribe personality to the impersonal machine of determinism no, not really.

Quote:

Conversely, if all is chaotic, then how could one argue an omnipotent design? Intended chaos?

The choices of God are not for mortals to understand.

(It is not possible to understand random choice).

[Dr Adequate]

Originally Posted by recursive prophet

As you seem to have a philosophical bent, here's a question you might be able to answer for me. I've oft seen it mentioned that random evolution is the creationist argument, and the logic escapes me.

The logic behind saying that this is the creationist argument is that this is, in fact, what creationists argue.

Quote:

If in fact our universe is deterministic, wouldn't this help make the case for a 'determiner?'

No. But since it appears not to be, your question is irrelevant.

Quote:

Conversely, if all is chaotic, then how could one argue an omnipotent design? If its all random, for what need have we gods?

You might be able to base an anti-theistic argument on the proposition that the universe is entirely random; but this would not be a reason for adopting the premise --- which is not, as it happens, true.

"A => B , I want to prove B, therefore A" is not a valid inference.

[recursive prophet]

Neither red nor blue, but green, eh Doc? Well, that seems to be the consensus opinion, though we have spent a lot of time arguing about shades. Many pages later nothing seems really resolved, so it appears you were right in your first post here about the futility of the subject.

At least on RDF Dawkins has finally entered the fray directly, and if he responds to susu's latest rebuttal I'll post it here. Otherwise, looks like we're the 'last men standing' here Doc, and the topic is done as we both now agree it's a cul de sac. Such is life. All we can learn leads to what we can not know.

[Walter Wayne]

Originally Posted by Dr Adequate

Ye have sown the wind ...

And have reaped a single drive-by post.

Originally Posted by recursive prophet

At least on RDF Dawkins has finally entered the fray directly, ...

I must be having a bad brain day; couldn't find the post by Dawkins in the thread.

Walt