What evidence is there for evolution being non-random?

[articulett]

I earlier stated ...

And articulett, I don't know why you keep talking about summing up evolution as random. No one here is trying to do that. We are merely stating one characteristic of it. I certainly wouldn't sum up my computer as "digital", but it is indeed one quality it has.

Walt

Actually, Mijo is-- and Behe does too. The majority seem to agree that it is generally a 2 part process--one fairly random-- and one that "selects" from the organisms built from the randomness.

Mijo's question is weird. It's like asking what is the evidence for Poker being non-random? He has concluded there is no evidence for any part of evolution being "non-random" per his definition of random-- he has done a sort of tortured semantic game and loose definition usage to call natural selection random. He is just trying to say that scientists really do think this all happened randomly-- which makes creationists leap to the 747 analogy. As long as they and their cohorts, don't understand how natural selection brings the appearance of design, then it all looks so unlikely that a designer must be involved. However, once you understand how natural selection makes things look designed that are not-- an intelligent designer seems superfluous. That's the whole wedge strategy in a nutshell.

Mijo's goal in every one of his posts is to ensure himself that creationists are correct in their silly strawman version of evolution. He will not ever convey natural selection is a simple and comprehensible way like the majority of people here on this forum have done-- like Dawkins did... like the peer reviewed paper did. He uses the word "random" in a way that no scientifically credible person does.

What do you think he's asking and saying? And why do you think he's bent on concluding that "there is no evidence for evolution being non-random". Truly, that statement is about as meaningful as saying "there is no evidence for Poker being non-random". He's doing the mealy mouthed semantic creationist tap dance where words are used to say nothing at all. He might fool you. He sure as hell doesn't fool most of us.

 

[mijopaalmc]

Dumbledore-

Would you like to explain how any of articulett's examples preclude evolution from being a stochastic process?

 

[Dumbledore]

Dumbledore-

Would you like to explain how any of articulett's examples preclude evolution from being a stochastic process?

Thats interesting, I noticed that you are answering my question with a question. Well I will actually answer your question, can you answer mine??? My guess would be that natural selection keeps evolution from being a stochastic process. While natural selection does involve randomness and probabilities it is not an stochastic process in of itself. Now multiple examples of evidence of this has been provided to you through this thread and others you have engaged in. Why do you continue the assertion that evolution is a stochastic process when you have be shown that natural selection by defination is not an stochastic process with multiple links to evidence that shows that natural selection is not a stochastic process?

Getting back to my question that you have thus far refused to answer. Why does the mainstream defination of evolution not regard evolution as a stochastic process. And an even early question I asked that you did not answer. If you are not convinced by the evidence we have posted why don't you find a friendly Phd in the subject and have them explain the technical details of the matter to you?

Furthermore you complain of being labeled a creationist when you claim that evolution is a stochastic process. It is not because of your claim that you get labeled as a creationist. It is because you act like a creationist or religious fundie, you ignore evidence and questions you don't like, and you blast anyone who does not agree with you as ignorant or misleading. Maybe if you don't like being labeled as a creationist you shouldn't act like one. A good place to start would be actually responding the questions that have been posed to you and the evidence that has been presented to you. But of course if you want to keep chanting your belief without listening to anyone else you can, just don't bother complaining that people label you as a creationist we you act just like one!

 

[mijopaalmc]

Thats interesting, I noticed that you are answering my question with a question. Well I will actually answer your question, can you answer mine??? My guess would be that natural selection keeps evolution from being a stochastic process. While natural selection does involve randomness and probabilities it is not an stochastic process in of itself. Now multiple examples of evidence of this has been provided to you through this thread and others you have engaged in. Why do you continue the assertion that evolution is a stochastic process when you have be shown that natural selection by defination is not an stochastic process with multiple links to evidence that shows that natural selection is not a stochastic process?

Getting back to my question that you have thus far refused to answer. Why does the mainstream defination of evolution not regard evolution as a stochastic process. And an even early question I asked that you did not answer. If you are not convinced by the evidence we have posted why don't you find a friendly Phd in the subject and have them explain the technical details of the matter to you?

Furthermore you complain of being labeled a creationist when you claim that evolution is a stochastic process. It is not because of your claim that you get labeled as a creationist. It is because you act like a creationist or religious fundie, you ignore evidence and questions you don't like, and you blast anyone who does not agree with you as ignorant or misleading. Maybe if you don't like being labeled as a creationist you shouldn't act like one. A good place to start would be actually responding the questions that have been posed to you and the evidence that has been presented to you. But of course if you want to keep chanting your belief without listening to anyone else you can, just don't bother complaining that people label you as a creationist we you act just like one!

Once again, Dumbledore, the bias in natural selection does not make evolution non-random; if more than one outcome exists for the same set of initial conditions (i.e., the survival or death of organisms with the same set of genes that contribute to the reproductive success of said organisms in a given environment), then the system is inherently stochastic. It is this fundamental indeterminacy of the result on the initial conditions that articulett and seemingly you seem to grasp.

 

[Dumbledore]

Once again, Dumbledore, the bias in natural selection does not make evolution non-random; if more than one outcome exists for the same set of initial conditions (i.e., the survival or death of organisms with the same set of genes that contribute to the reproductive success of said organisms in a given environment), then the system is inherently stochastic. It is this fundamental indeterminacy of the result on the initial conditions that articulett and seemingly you seem to grasp.

Well I suggest you find a Phd that does grasp the argument, they would be happy point out why this is not true, since the mainstream defination of evolution does not regard it as a stochastic process, I think that is my tenth time typing that last statement.

Nor are you dealing with the same set of initial conditions, species within themselves will have different genes, thus lending to the different success of different members of that species. As was shown with the link relating to the butterflys showing how natural selection works.

In fact your entire argument is faulty if you assume set initial conditions,i.e. the same genes for every member of a species, natural selection by definition shows us that the initial conditions between members in a species, their genes are not the same for all the members, hence some members succeed because they have the right genes and others do not, because they don't have the right genes.

Furthermore reproductive success within an environment does not guarantee a species or groups success. But if you are not convinced, you should seek a Phd, they would be able to hash out the nitty gritty to you. By the way you still have not answered my question, why in the mainstream defination of evolution is evolution not regarded as a stochastic process?

 

[mijopaalmc]

By the way you still have not answered my question, why in the mainstream defination of evolution is evolution not regarded as a stochastic process?

Beg the question much?

Just because articulett says that no-one in the mainstream says that evolution can be a stochastic process doesn't mean that it's true. In fact is a myth.

Here are two reviews that show that a vast number of scientist over 85 years have no problem assuming that evolution is a stochastic process:


Transition between Stochastic Evolution and Deterministic Evolution in the Presence of Selection: General Theory and Application to Virology
Stochastic Models of Evolution in Genetics, Ecology and Linguistics

 

[Dumbledore]

Beg the question much?

Just because articulett says that no-one in the mainstream says that evolution can be a stochastic process doesn't mean that it's true. In fact is a myth.

Here are two reviews that show that a vast number of scientist over 85 years have no problem assuming that evolution is a stochastic process:


Transition between Stochastic Evolution and Deterministic Evolution in the Presence of Selection: General Theory and Application to Virology
Stochastic Models of Evolution in Genetics, Ecology and Linguistics

Sorry but neither of these links show that these scientists or anyone else defines evolution itself as a stochastic process. The use of stochastic models does not mean that evolution itself is defined as a stochastic process in its entirety. If that was so the scientists might write, a shift in thinking of evolution as an stochastic process, they do not. Again the use of stochastic models does not mean that evolution is a stochastic process, it only means that stochastic models are useful in examining parts of evolution and its effects.

 

[Walter Wayne]

Sorry but neither of these links show that these scientists or anyone else defines evolution itself as a stochastic process. The use of stochastic models does not mean that evolution itself is defined as a stochastic process in its entirety. If that was so the scientists might write, a shift in thinking of evolution as an stochastic process, they do not. Again the use of stochastic models does not mean that evolution is a stochastic process, it only means that stochastic models are useful in examining parts of evolution and its effects.

It is true that a stochastic model does not necessarily mean a stochastic system. However, we know some of the mechanisms for mutation, and that mutation is in fact random. Even if I stipulate for this argument that selection is deterministic, what does that mean as regards to whether the combination of the mutation and selection is random?

The fact is that a variety of mutations can pass the selection process, and that not all of these mutations will necessarily happen. As a result, even if we regard selection as 100% determistic, the system comprised of mutation and selection is random as you can not predict what will come out of it. Applying a deterministic process to random data can produce random results (and in most natural system it will). The filter would have to have very specific properties in order to produce non-random outputs from random inputs, and natural selection doesn't have these.

The argument that evolution is non-random, or not a stochastic process, is completely untenable.

Walt

P.S. I can see where someone might attempt to defend the idea that natural selection isn't stochastic or random, and the argument is more complex. However, if some tells you evolution is not random, they fail to understand some very basic concepts about random systems.

 

[Paul C. Anagnostopoulos]

As a result, even if we regard selection as 100% determistic, the system comprised of mutation and selection is random as you can not predict what will come out of it.

Hang on now. Of course you can predict what will come out of it: Organisms that are adapted to their environment. Perhaps you mean that you can't predict the specific features of organisms. But how specific are we talking? I predict that if you feed enough of an antibiotic to some germs, you will eventually get ones that are resistant.

I agree that evolution is a stochastic process, but, just like the weather, you can still predict it.

~~ Paul

 

[mijopaalmc]

Hang on now. Of course you can predict what will come out of it: Organisms that are adapted to their environment. Perhaps you mean that you can't predict the specific features of organisms. But how specific are we talking? I predict that if you feed enough of an antibiotic to some germs, you will eventually get ones that are resistant.

I agree that evolution is a stochastic process, but, just like the weather, you can still predict it.

I'm sorry isn't the statement "orgnaisms will adapt to their environment (or not and become extinct)"* is a prediction of evolution sort of like saying "there will be weather" is a prediction of meteorology?

*The parenthetical is necessary because not all organisms or species survive (e.g., dinosaurs didn't adapt to whatever environmental change occurred and the end of the Cretaceous).

 

[Dumbledore]

It is true that a stochastic model does not necessarily mean a stochastic system. However, we know some of the mechanisms for mutation, and that mutation is in fact random. Even if I stipulate for this argument that selection is deterministic, what does that mean as regards to whether the combination of the mutation and selection is random?

The fact is that a variety of mutations can pass the selection process, and that not all of these mutations will necessarily happen. As a result, even if we regard selection as 100% determistic, the system comprised of mutation and selection is random as you can not predict what will come out of it. Applying a deterministic process to random data can produce random results (and in most natural system it will). The filter would have to have very specific properties in order to produce non-random outputs from random inputs, and natural selection doesn't have these.

The argument that evolution is non-random, or not a stochastic process, is completely untenable.

Walt

P.S. I can see where someone might attempt to defend the idea that natural selection isn't stochastic or random, and the argument is more complex. However, if some tells you evolution is not random, they fail to understand some very basic concepts about random systems.

My point is that the mainstream definition of evolution does not define it as a stochastic process. Here are links showing that evolution is not a random process.

http://www.pbs.org/wgbh/evolution/library/faq/cat01.html#Q07
http://evolution.berkeley.edu/evolibrary/article/0_0_0/misconcep_05
http://www.talkorigins.org/faqs/faq-misconceptions.html

If you are not convinced by them seek others, if you are not convinced by others you find, find a Phd in the subject. Repeat till convinced, if still not convince, seek Phd in subject and seek a redefinition of evolution through research.

My point is that I don't have to prove that evolution is not a stochastic process, the majority of scientists do not view it as so. The burden is on you to show that evolution is a stochastic process! In short you have to prove why evolution is a stochastic process when most scientists do not believe it is so. In short, why are you right, no Phd's mentioned, and they are wrong, plenty of Phd's to back up their claims.

 

[Paul C. Anagnostopoulos]

I'm sorry isn't the statement "orgnaisms will adapt to their environment (or not and become extinct)"* is a prediction of evolution sort of like saying "there will be weather" is a prediction of meteorology?

No, "there will be change" would be more analogous to "there will be weather." And change is all we could expect if evolution was nothing but a random process with respect to the environment, as a coin toss is. But evolution is not random with respect to the environment, so we can predict that the change will go in the direction of making organisms adapt to the environment (or go extinct).

This is why "evolution is random" is not a helpful description, even if technically correct.

~~ Paul

 

[Paul C. Anagnostopoulos]

My point is that I don't have to prove that evolution is not a stochastic process, the majority of scientists do not view it as so. The burden is on you to show that evolution is a stochastic process! In short you have to prove why evolution is a stochastic process when most scientists do not believe it is so. In short, why are you right, no Phd's mentioned, and they are wrong, plenty of Phd's to back up their claims.

Let me quote the Penguin Dictionary of Mathematics:

stochastic process. A random process. Common usage excludes essentially deterministic processes, which are subject only to random errors.

There you have it, my friends. If you think that mutations are simply "random errors," then you are justified in asserting that evolution is not a stochastic process. If, on the other hand, you maintain that mutation is such an integral part of the process that dismissing it* as trivial "random error" is unjustified, then evolution is a stochastic process.

Once again, perhaps a bit more explanation than "evolution is random" would help the uninitiated.

~~ Paul

* As, for example, you would dismiss the occasional computer memory parity error as not an integral part of computing, and therefore claim that computers are deterministic.

 

[steenkh]

There you have it, my friends. If you think that mutations are simply "random errors," then you are justified in asserting that evolution is not a stochastic process.

Nope, you got that wrong. It is the selection process that is the deterministic process, not the mutations.

 

[mijopaalmc]

Nope, you got that wrong. It is the selection process that is the deterministic process, not the mutations.

Again, evidence?

You and other who have argued similarly have provided evidence that scientists say that evolution is not random but that doesn't make it non-random. In order to demonstrate that it is non-random you need to provide evidence that there is a collection of alleles the possessors of which always survive and never die and deviation from which guarantees death.

 

[Dumbledore]

Again, evidence?

You and other who have argued similarly have provided evidence that scientists say that evolution is not random but that doesn't make it non-random. In order to demonstrate that it is non-random you need to provide evidence that there is a collection of alleles the possessors of which always survive and never die and deviation from which guarantees death.

First of all we argue that evolution is not defined as a stochastic process by mainstream definitions because that is what mainstream scientists believe. If you want to prove otherwise it is up to you to prove other wise. Furthermore why do you feel the need to have technical discussions about evolution with people who are not experts in the field. It seems to me that if you wanted answers to your questions you would actually find experts in evolution, of course they might show how you are wrong!

 

[Ichneumonwasp]

I apologize if this is already covered since I got only this far on this page (I can't believe this thread is still rehashing the same arguments).

Once again, Dumbledore, the bias in natural selection does not make evolution non-random; if more than one outcome exists for the same set of initial conditions (i.e., the survival or death of organisms with the same set of genes that contribute to the reproductive success of said organisms in a given environment), then the system is inherently stochastic. It is this fundamental indeterminacy of the result on the initial conditions that articulett and seemingly you seem to grasp.

But that's just it. There is no fundamental indeterminacy, unless we speak of everything as being fundamentally indeterminate ala Heisenberg's Uncertainty Principle. Is that what you mean, that everything is fundamentally indeterminate?

This argument is semantic. For most purposes, random does not mean fundamentally indeterminate. It means that we cannot predict an outcome because we lack certain knowledge of the starting conditions. We do not speak of processes like natural selection as random because they tend to limit the possible outcomes. Anything that limits possible outcomes, whether or not the process itself has random elements to it (and selection does have such random elements, as Gould was always fond of pointing out -- the contingencies of life) we view as something quite different from random. The way we generally use the word "random" refers to the relative unpredictability of outcome -- we cannot predict which of the 6 possibilities a die roll will produce (we could actually if we had enough knowledge of the die and the person throwing it and the wind velocity and the angle of attack and the -- you get the picture). We cannot predict which mutations will arise in a given genome (we could if we had enough knowledge to watch the process unfold, except possibly for radiation induced mutations). But selection, which again has random elements, limits these possibilities to less than is likely in a state without selection.

We do not call the process of mutation and natural selection random for the same reason we do not call a loaded die when rolled a "random toss". The number of outcome possibilities has been limited. So selection, which has random elements to it, imposes order on the whole process and therefore makes the process less random (less unpredictable).

And the really funny thing is that selection can do this even when it is really random -- say a comet hitting the earth. Without lots of detailed knowledge we could never have predicted that, so it is a random element in selection. But this very random element seriously limits the outcome options, so imposes significant order (kills off all sorts of different possibilities leaving behind only limited survivors).

Let's get real when it comes to the "random" argument as it relates to evolution. It is an argument about order and disorder.

 

[Ichneumonwasp]

You and other who have argued similarly have provided evidence that scientists say that evolution is not random but that doesn't make it non-random. In order to demonstrate that it is non-random you need to provide evidence that there is a collection of alleles the possessors of which always survive and never die and deviation from which guarantees death.

What? How did this become the definition of non-random? I think one of the problems here is that we have a complex systems with many parts and you are trying to impose a single term on it.

If the universe worked as Newton thought, then it would be the case that the random/non-random distinction would be a mere semantic abstraction based on our level of knowledge, which would make this a trivial debate. If the universe works on a radical view of Uncertainty -- that all events are completely unpredictable at a sub-atomic level -- then everything is random and this is a trivial debate. If the unpredictable events "wash out" at higher levels of description so that the world appears like what Newton described, then this is a trivial debate.

At its core, the universe is either predictable or unpredictable. Either way, the use of random in these discussions refers to our level of knowledge, not a fundamental property of the universe. We use the word to refer to chaotic systems where outcomes are greater than one. When outcomes are limited by some process we do not refer to that process as random whatever the mechanism by which the process works.

 

[Ichneumonwasp]

It is true that a stochastic model does not necessarily mean a stochastic system. However, we know some of the mechanisms for mutation, and that mutation is in fact random. Even if I stipulate for this argument that selection is deterministic, what does that mean as regards to whether the combination of the mutation and selection is random?

The fact is that a variety of mutations can pass the selection process, and that not all of these mutations will necessarily happen. As a result, even if we regard selection as 100% determistic, the system comprised of mutation and selection is random as you can not predict what will come out of it. Applying a deterministic process to random data can produce random results (and in most natural system it will). The filter would have to have very specific properties in order to produce non-random outputs from random inputs, and natural selection doesn't have these.

The argument that evolution is non-random, or not a stochastic process, is completely untenable.

Walt

P.S. I can see where someone might attempt to defend the idea that natural selection isn't stochastic or random, and the argument is more complex. However, if some tells you evolution is not random, they fail to understand some very basic concepts about random systems.

As Paul said, yes, we can know something about what will drop out. If you mean that we cannot know all the details of the organisms that will drop out, no one argues that (at least no one that I know). The argument is really over the process being seen as completely chaotic since that is the straw man protrayal of the anti-evolution crowd. Since selection limits the possible outcomes (some of the random mutations do not survive), the process is not completely chaotic -- there is some order imposed upon it. If you use the technical definition of random, then the process is random because it has random elements. The important point for discussing this with creationists, however, is the element that imposes some sense of order to limit the outcomes so that total chaos does not reign. Keep in mind that their characterization of the process is "total chaos, so order is not possible". It is unfortunate, I think, that we find it difficult to speak in precise terms and use the technical definition for random in this sort of discussion because one group tends to take advantage of the language being used. I'm fully with you guys who want to teach them how to use such language properly, but I'm still not holding out much hope.

 

[Walter Wayne]

As Paul said, yes, we can know something about what will drop out. If you mean that we cannot know all the details of the organisms that will drop out, no one argues that (at least no one that I know). The argument is really over the process being seen as completely chaotic since that is the straw man protrayal of the anti-evolution crowd. Since selection limits the possible outcomes (some of the random mutations do not survive), the process is not completely chaotic -- there is some order imposed upon it. If you use the technical definition of random, then the process is random because it has random elements. The important point for discussing this with creationists, however, is the element that imposes some sense of order to limit the outcomes so that total chaos does not reign. Keep in mind that their characterization of the process is "total chaos, so order is not possible". It is unfortunate, I think, that we find it difficult to speak in precise terms and use the technical definition for random in this sort of discussion because one group tends to take advantage of the language being used. I'm fully with you guys who want to teach them how to use such language properly, but I'm still not holding out much hope.

But most of them understand that a die rolled on a flat hard surface doesn't come up as a edge, or a corner. And they still call that random. They call mutation random, even though the outcomes of that is limited by the previous material. If you look through random processes, even those that laymen call random, the vast majority of them do have limited outcome. The problem may partly be misunderstanding the term, but a large part of it seems to be a block when it comes to this particular process.

Walt