Natural selection

[Wowbagger]

So why the adamant refusal to consider that randomness could be a fundamental and irremovable characteristic of evolution by natural selection?

I don't refuse to admit that it might be an irremovable characteristic of our models.

But, no matter how you define random, evolution might NOT really be fundamentally random in that way.

A model that works well in predicting the outcome of a process may not actually be describing how the process proceeds, but the fact that it works well should be an indicator that it is at least possible (if not extremely likely) that the model is actually be describing how the process proceeds.

Sure it is possible. But, the general trend seems to be that, the more details we learn about life, the better we are able to predict stuff about it.

This general trend is consistent with the idea that life is not random, in any way you define the word, at the scales nature works with.

If we zoom into life, long enough, we could hit the quantum level, where randomness has been demonstrated as fundamental (near as we can tell). But, it should be noted that, as we zoom out again, that quantum randomness gets averaged out, so its effects are rendered miniscule, if they even effect anything at all.

That is how we can make discoveries like these:
http://www.sciencedaily.com/releases/2007/11/071119123929.htm

By the way, many to the of the foundational thinker who constructed the Modern Synthesis of Evolutionary Theory (e.g, Wright, Fisher and Haldane) had no trouble thinking of and referring to evolution by natural selection as a stochastic process as they were both gifted mathematicians and talented biologists. Now, there is the legitimate objection that they weren't actually calling the process itself stochastic but merely considering it to be stochastic to ease the modeling process;

I agree. I find it funny how all that helps re-affirm my case.

however, the fact that such great minds thought of evolution by natural selection as stochastic seems to imply that such an idea deserves more than just the out-of-hand rejection that it receives to day form people whose mathematical credentials are suspect.

So now we are clear on the matter, I hope.

Mathematical credentials help, but cannot, alone, make discoveries about the Universe. Hence the comment in my previous point.

 

[jimbob]

This general trend is consistent with the idea that life is not random, in any way you define the word, at the scales nature works with.

If we zoom into life, long enough, we could hit the quantum level, where randomness has been demonstrated as fundamental (near as we can tell). But, it should be noted that, as we zoom out again, that quantum randomness gets averaged out, so its effects are rendered miniscule, if they even effect anything at all.

Except in chaotic systems, where they would have sufficient effect to actually alter weater significantly enough to affect whether a particular organism would survive or not. A "random" breath of wind indicating the presence of a lion to its prey...

A "random" breath of wind affecting location of a tree seed...

It could be averaged out, but could still be random, and need stats to make a realistic analysis work.

 

[Taffer]

I'm not being ignored here mijopaalmc and Walter, am I?

 

[Wowbagger]

A "random" breath of wind affecting location of a tree seed...

Is the wind truly "random" or are we humans forced to model it that way, because we are not going to be privy to all the countless variables involved?

You don't need to answer. We end up going off on needless philosophical and semantic journeys, about this.

But, it is sufficient to say that Evolution is non-random enough, that its power can help explain why the poor tend to have more children than the rich (on average).

 

[jimbob]

Wowbagger, I'd say that weather *is* random, because of the nonlinear effect of quantum-level variations on a chaotic system.

When I was at university, I think the calculations were that a 6-week forcast would need to account for quantum-level events (whether or not a particular air molecule was ionised by a radioactive decayu event, for example)...


But, as you say, that is a partial derail for this thread.

 

[mijopaalmc]

I don't refuse to admit that it might be an irremovable characteristic of our models.

But that was not my point. I was saying that people seem to be deliberately ignoring the fact that the actual process of evolution by natural could be stochastic. The rest of your post, especially then next sentence clearly demonstrates it.

But, no matter how you define random, evolution might NOT really be fundamentally random in that way.

Wow, that is probably the most stunningly blind statement you have made in this thread.

Sure it is possible. But, the general trend seems to be that, the more details we learn about life, the better we are able to predict stuff about it.

This general trend is consistent with the idea that life is not random, in any way you define the word, at the scales nature works with.

If we zoom into life, long enough, we could hit the quantum level, where randomness has been demonstrated as fundamental (near as we can tell). But, it should be noted that, as we zoom out again, that quantum randomness gets averaged out, so its effects are rendered miniscule, if they even effect anything at all.

Again, you, like cyborg and Taffer, seem to be ignoring the fact that I have given and am using the technical definition of "random" (i.e., [o]f or relating to a type of circumstance or event that is described by a probability distribution) and not any other definition. Thus our ability to predict things about life has nothing to do with whether evolution by natural selection is random, as one can use probability to make extremely accurate conditions regardless of whether a system is actually random or not.

That is how we can make discoveries like these:
http://www.sciencedaily.com/releases/2007/11/071119123929.htm

Yet again, presenting this article as evidence that evolution is not random, especially when "random" has been clearly and repeatedly defined for you as "[o]f or relating to a type of circumstance or event that is described by a probability distribution", is equivocation pure and simple. Yes, the article does quite clearly provide evidence that evolution by natural selection is not "random" in so far as it is not "without direction" but that does not answer the question of how it is not "random" in so far as it is "[o]f or relating to a type of circumstance or event that is described by a probability distribution".

I agree. I find it funny how all that helps re-affirm my case.

Seeing only what you want to see?

So now we are clear on the matter, I hope.

Mathematical credentials help, but cannot, alone, make discoveries about the Universe. Hence the comment in my previous point.

You're missing the point yet again. The wast majority of evolutionary biologists will rail against the randomness of evolution by natural selection by saying that it is "the exact opposite of random chance" and the go on to say something to the effect that individuals within a species who are more adept at accessing the resources within their environment are more likely to survive and reproduce, which in turn favors the persistence of better-adapted forms. Now, this may be a good explanation of why evolution by natural selection is not "random" in the sense of being "eqiuprobable" or "without direction", but it is a self-contradictory statement when you define random as "[o]f or relating to a type of circumstance or event that is described by a probability distribution". Since that is the technical definition of random, it seems reasonable to question the mathematical acumen and credentials of a person who make the above two statements.

 

[Wowbagger]

But that was not my point.

This is ridiculous! What more do you want?! You already agreed with my point. You seem to accept that models are called stochastic, for ease of use. What else is there to do?

Wow, that is probably the most stunningly blind statement you have made in this thread.

Maybe the wording could have been better? Perhaps it might be better to summarize that point as: Nature scoffs at your definitions.

Mathematics is important. Don't get me wrong. Stochastic models are useful, no doubt. But, the general attitude of scientists, who study these things, is that they are only models. Don't get all bent out of shape on this. It's not worth it.

Thus our ability to predict things about life has nothing to do with whether evolution by natural selection is random, as one can use probability to make extremely accurate conditions regardless of whether a system is actually random or not.

Then you were fighting the wrong battle, the whole time.

The randomness I was fighting against was the type where predictions would be hopeless. The kind Creationists like to accuse Evolution of.

Next time pay more attention.

The wast majority of evolutionary biologists will rail against the randomness of evolution by natural selection by saying that it is "the exact opposite of random chance" and the go on to say something to the effect that individuals within a species who are more adept at accessing the resources within their environment are more likely to survive and reproduce, which in turn favors the persistence of better-adapted forms.

No, it is you who misses the point of this language, quite severely.

When the biologists say "evolution is the opposite of random chance", they mean evolution is NOT the form of randomness where predictions would be useless and hopeless.

When those same biologists say a species is "more likely to survive", they mean that the survival was pretty much pre-determined by nature. Though, since our information is imperfect, we have to resort to using a little bit of probability in the general model. (Especially since there are other factors involved in survival than the few discussed at the time.)

If we had (hypothetically) perfect knowledge of life, it would no longer be a matter of "more likely", we would know precisely: THIS species will survive and thrive. THIS one will not. Etc.

The general trend is that, the more precision we learn about life, the less of a role guesswork and statements such as "more likely" tend to play.

 

[Belz...]

Here again you show your mastery of equivocation. "Random" as I am using does not have anything to do with causal relationship; it is purely based in probabilities.

How can it have nothing to do with causality ?

 

[Belz...]

Actually, the wavefunction describes all that it is possible to know about the ensemble. It just happens that what is possible to know come in the form of a range of values (i.e., a probability distribution) rather than a single exact value.

If you only know where it COULD be then you don't know where it IS, do you ?

 

[Walter Wayne]

I'm not being ignored here mijopaalmc and Walter, am I?

Not sure what you want me to address. If you want my definition of randomness I contend that evolution is random in the technical sense (future state unknown based on present state), and for just about every laymans definition of random; the sense of every possibility being equal probability being the only one I think doesn't apply to evolution.

Edited to add: Earthborn covered several reasons why natural selection is random in the layman sense, even if no technically random process is found.

If there was something else you wanted me to address in particular, ask away.

Walt

 

[Walter Wayne]

Causality:

A die roll is caused by the person tossing the die. Yet the die roll is random (in layman terms at the very least).

An electron dropping to a lower orbit causes emission of a photon, where that photon goes is random. It may cause excitation of a different random electron. A series of random events linked causally.

Also, Belz and Cyborg, you seem to still confuse quantum indeterminancy with measurement inaccuracy. This was discussed in a thread a while back.

Walt

 

[Taffer]

Not sure what you want me to address. If you want my definition of randomness I contend that evolution is random in the technical sense (future state unknown based on present state), and for just about every laymans definition of random; the sense of every possibility being equal probability being the only one I think doesn't apply to evolution.

Edited to add: Earthborn covered several reasons why natural selection is random in the layman sense, even if no technically random process is found.

If there was something else you wanted me to address in particular, ask away.

Walt

Nope, that was the only thing.

I'm just after a concise definition of randomness. Such as "Randomness is any event that is acausal". Then, include that into your line of reasoning that evolution is random, for example "evolution is an acausal even, therefore evolution is random".

I know the answer to this question is to be found somewhere with this thread, but I just hope you can quickly summerize it for me.

ETA: Please bear in mind that that I do not consider "so complex we cannot understand all the processes involved" to be "technically random".

 

[cyborg]

A series of random events linked causally.

Fail:

It may cause excitation of a different random electron.

In other words there is no cause as to whether it is one electron or the other -> acausal.

Also, Belz and Cyborg, you seem to still confuse quantum indeterminancy with measurement inaccuracy.

Nope.

 

[jimbob]

Nope, that was the only thing.

I'm just after a concise definition of randomness. Such as "Randomness is any event that is acausal". Then, include that into your line of reasoning that evolution is random, for example "evolution is an acausal even, therefore evolution is random".

I know the answer to this question is to be found somewhere with this thread, but I just hope you can quickly summerize it for me.

ETA: Please bear in mind that that I do not consider "so complex we cannot understand all the processes involved" to be "technically random".

If identical conditions do not lead to the same outcome every time, then I would consider that to be a working definition of "random".

I would argue that a quantum decay event is the archetypal "random event".

In the context of evolution, I would argue that because weather is a highly nonlinear system, quantum events can be magnified to have a significant effect on the weather. If this is the case, then there is going to be a random element in natural selection. I would also argue that the relationships between organisms is even more nonlinear than the weather, and a chance mutation happening befoore anonther could alter the selective pressures on other organisms within the ecosystem, and thus alter the "direction" of evolution in the ecosystem.

This means that should one have the luxury of creating identical universes just before the KT impact, the course of evolution in each of these initially identical universes would diverge.

Given the low chance of any individual organism managing to have reproducing offspring, I would contend that there was noting inevitable about the emergence of hominids, until some time after the last common ancestor with apes.

Evoulution would happen in all the other cases, it is just that the species mix, and indeed occupied niches would probably differ in each case. There is nothing special abut the emergence of humanity, except to us...

 

[rocketdodger]

wtf why does every evolution thread turn into this stupid "your definition of random is not the right definition" nonsense?

You realize you are arguing about nothing of any concern, right?

 

[jimbob]

wtf why does every evolution thread turn into this stupid "your definition of random is not the right definition" nonsense?

You realize you are arguing about nothing of any concern, right?

I am talking to Taffer, walter wayne and wowbagger, as I (maybe foolisly) think that there could be a consensus there...

Wowbagger, I woul dcertainly agree that we can make predictions based on evolutionary theory.

For example: Flight and sight are obviously very likely to arise.

Birds (especially herbivorous birds) will tend to lose flight if isolated somewhere with no predators, as the energy cost is large...

I would take issue with the idea that the theory of Evolution could explain the following:

But, it is sufficient to say that Evolution is non-random enough, that its power can help explain why the poor tend to have more children than the rich (on average).

Until the advent of birth control, I would contend that the opposite was probably true, especially surviving offspring

(the name "Fitzroy", being an example of this, or the Asian y-chromosome, that seemed to originate form a single ancestor about 700 years ago...)

It can explain why there are more cod spawn per parent than baby orcas.

 

[Walter Wayne]

wtf why does every evolution thread turn into this stupid "your definition of random is not the right definition" nonsense?

You realize you are arguing about nothing of any concern, right?

Because some people debating creationists, and talking in front of or to people who wish to learn like to cram "it's not random" down there throat. Which leads to misinformed people.

If I didn't see the trend of some ranting "but it isn't random" so often I wouldn't care.

 

[Taffer]

If identical conditions do not lead to the same outcome every time, then I would consider that to be a working definition of "random".

Agreed.

I would argue that a quantum decay event is the archetypal "random event".

Also agreed.

In the context of evolution, I would argue that because weather is a highly nonlinear system, quantum events can be magnified to have a significant effect on the weather. If this is the case, then there is going to be a random element in natural selection. I would also argue that the relationships between organisms is even more nonlinear than the weather, and a chance mutation happening befoore anonther could alter the selective pressures on other organisms within the ecosystem, and thus alter the "direction" of evolution in the ecosystem.

If we assume weather to be affected at a significant level by quantum events, I would agree with you. I also agree that interorganism interactions are very complex, and specific mutation events are boardering on quantum randomness. Of course, we don't know that for certain, but so far that is how it looks.

This means that should one have the luxury of creating identical universes just before the KT impact, the course of evolution in each of these initially identical universes would diverge.

Agreed.

Given the low chance of any individual organism managing to have reproducing offspring, I would contend that there was noting inevitable about the emergence of hominids, until some time after the last common ancestor with apes.

Agreed.

Evoulution would happen in all the other cases, it is just that the species mix, and indeed occupied niches would probably differ in each case. There is nothing special abut the emergence of humanity, except to us...

Yep, couldn't put it better myself.

Walter Wayne, this is a perfect example of what I asked for.

 

[mijopaalmc]

Where exactly a people getting the definition of "random" as "acausal"?

It does not occur in any dictionary I have checked (including the Oxford English Dictionary). Yes, I realize that dictionaries only describe how a word is defined, but one would think that such a description would be found in a dictionary if the word were used in such a way.

 

[cyborg]

Given the low chance of any individual organism managing to have reproducing offspring, I would contend that there was noting inevitable about the emergence of hominids, until some time after the last common ancestor with apes.

Selection can only occur on the current generation so all the historical possbilities that could have happened but didn't are rather irrelevant.