Register to reply

Random evolution

by binbots
Tags: evolution, random
Share this thread:
zomgwtf
#37
Apr3-10, 11:13 AM
P: 500
Quote Quote by sganesh88 View Post
Tkjtkj can i have your response for my arguments? Am i going wrong somewhere?
You are changning the definitions of the words you are using, such as 'evolution' and 'random' in order to fit a preconceived notion that evolution IS random.

It truly is NOT random, it just HAPPENS naturally.

There is no CHANCE involved in natural selection. It is either you reproduce or you don't and it is assumed that those who have the opportunity to reproduce were BETTER SUITED for mating in an ENTIRE POPULATION. Why? Because they DID reproduce. This means that THEIR genes are inherited, rinse and repeat. Sure in HINDSIGHT you can say 'oh it was shear luck that this organism with this gene mated with this one and had x children with said gene' but it WAS NOT LUCK. IT WAS NATURE. Over time these changes can be seen and THAT is what is called evolution. There is no dice rolling in this to decide who mates and who doesn't.

Mind you this is quite a dumbed-down version.
sylas
#38
Apr3-10, 12:16 PM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by zomgwtf View Post
You are changning the definitions of the words you are using, such as 'evolution' and 'random' in order to fit a preconceived notion that evolution IS random.

It truly is NOT random, it just HAPPENS naturally.

There is no CHANCE involved in natural selection. It is either you reproduce or you don't and it is assumed that those who have the opportunity to reproduce were BETTER SUITED for mating in an ENTIRE POPULATION. Why? Because they DID reproduce. This means that THEIR genes are inherited, rinse and repeat. Sure in HINDSIGHT you can say 'oh it was shear luck that this organism with this gene mated with this one and had x children with said gene' but it WAS NOT LUCK. IT WAS NATURE. Over time these changes can be seen and THAT is what is called evolution. There is no dice rolling in this to decide who mates and who doesn't.
This doesn't fit my notions of random, or chance, or evolution, or selection.

Chance is critical in the theory of evolution, and specifically in population genetics. For example, you can give probabilities for a particular mutation to become fixed or eliminated in a population. There's a definite and quite substantial, in fact, chance that a beneficial mutation will be eliminated. That's because natural selection is about amplifying the small changes in probabilities that arise in all the random chances of life for an individual with a given mutation.

For example, have a look at Population genetics: a concise guide, by John H. Gillespie (2004). I found that reference just with a quick google because I knew pretty much any basic reference would tell the same story. Note that there's an appendix on probability. It's essential to figuring out how evolution works.

Another important feature of evolution is called "genetic drift". This is basically change which has negligible consequences for an organism, so there's no natural selection involved. It's still evolution, by biological definitions.

A really good resource for learning more about evolutionary theory is the talkorigins archive. A sample article there is Random Genetic Drift , by Professor Larry Moran. This is a website intended to give basic information for interested readers; and I think it is an excellent starting point. Full disclosure; I've long been involved personally in talkorigins, which is current moribund after a hack attempt probably from creationists. (They really don't like it.) But I think it will be back up and working again sometime soon with a whole pile of new features.

So I disagree. Evolution truly IS random. But random is not the same as a coin flip. Evolution works its wonders by biasing the dice. And it is amazing what you can achieve with a small bias in a random chances of life. I'm also an ex-blackjack card counter. I was able to get maybe a 1% edge over the casino until I was caught and banned and had to find another hobby. Was it random. Sure. But will the casino always make money? Sure. They make sure they have the edge, and that's all they need.

Cheers -- sylas

PS. Your caps lock key seems to be sticking.

PPS. I'm using the conventional biological definition of evolution as any change in the distributions of heritable characteristics within a population over time.
zomgwtf
#39
Apr3-10, 01:02 PM
P: 500
Quote Quote by sylas View Post
PPS. I'm using the conventional biological definition of evolution as any change in the distributions of heritable characteristics within a population over time.
The genetic changes that occur are definitely random yes... no one has said otherwise. Evolution however is not individual genetic changes in an organism but genetic change in an entire population of species. It's not by chance that the organism gets to mate and continue the inheritance, it is something that occurs naturally for various reason.

Only in hindsight can we look back and say 'well it's lucky that humans evolved larger brains when they could have just as easily stayed with more primative brains,' the thing is though that it was not by chance but by natural selection and other mechanisms which I don't feel like getting into in this particular thread.
zomgwtf
#40
Apr3-10, 01:06 PM
P: 500
Quote Quote by sylas View Post
This doesn't fit my notions of random, or chance, or evolution, or selection.
Random- governed by or depending on chance;

Chance- luck: an unknown and unpredictable phenomenon that causes an event to result one way rather than another

Evolution- (biology) the sequence of events involved in the evolutionary development of a species or taxonomic group of organisms
OR
change in the inherited traits of a population of organisms through successive generations

Both meaning the same thing. Of course randomness has effects on Evolution but that is far from being the same as saying Evolution is random. Period.
Jon Richfield
#41
Apr3-10, 01:22 PM
P: 258
Hi zomgwtf,
Thanks for the call.
>I have to disagree with what you have posted.<
“No cow is like a horse.
No horse is like a cow.
That’s one similarity anyhow.”
Ogden Nash
So we can be equally comfortable in the closeness of our agreement. Now…
>Genetic variation is possibly the only 'random' event that leads to evolution...<
Not at all very possibly! For one thing, I rather fancy I pointed out the contingent nature of randomness, though I did not spell it out because it is an old topic. Consider: suppose I were to give you this string of digits to evaluate for randomness: 5517027618386062613313845 Perhaps you say: Ahah! It fails the test in such and such an attribute. Then I say: try this one: 7245870066063155881748815. You react in much the same way. OK, I say, you don’t think much of my random number generator, but let’s see you correlate the two. In fact, lets see you correlate a mathematically large expansion of those digits with any other sequence that is mathematically independent of them.
So far so trivial, but it leaves one of us with the problem of how one is to decide whether the numbers are random or not. Even if you do not fail to correlate either of the sequences with some function each, unless you can predict how the next digit of one sequence will correlate with the corresponding digit of the other without calculating both of these first, how can you say that they are not random relative to each other?
Now, as you have noticed, these two sample strings are in fact from highly non-random sequences, in the sense that they can be generated to arbitrary lengths by fairly short programs. In fact programming courses often include their generation as demonstration exercises. One comes from pi and the other from e. From another point of view however, they are highly random. From what external criterion that does has no isomorphism with either of them, would you undertake to predict their next digit, much less the mod 10 difference between the two?
What, you might ask in rebuttal, does that have to do with randomness in evolution? In particular what does it have to do with randomness other than in genetic variation in evolution?
Well, firstly, just how random is genetic variation? Not really very. There are all kinds of biases and constraints that make certain mutations likelier than others, and what is more, make the effect of certain mutations more relevant than others. To put it another way, we might not easily predict the next effective mutation, but we would hate to use spontaneous genetic change as a random number generator!
So genetic change is not as random as we might like, not as mathematicians or computer scientists anyway. But as biologists we might well argue that it is random enough for jazz. However, what about the other variables? What other variables are there? The first thing to bear in mind is that we are customarily very loose in our use of terms like “evolution”. Usually people say “evolution” when they mean “Darwinism” which is more like the effect on a population, of natural selection. Evolution, as I am sure you realise, is something more like the more or less systematic change of a population’s gene pool (the relative frequencies of given genetic attributes from generation to generation, if you like). But what is natural selection? It is (correct me by all means!) the relative frequency of successful reproduction of organisms of distinct heritably determined phenotypes in given ecological circumstances. A little more indirectly (or directly, if your name happens to be Richard) this translates into selection for given genetic attributes.
Well now, what can we do about predicting the nature of our ecology? Is it random? Is it??? If so, with respect to what? The genomes of the population? You might have some trouble denying that one! Let’s see you justify your prediction of the correlation between whether our wolf cub gets poisoned by a touch of botulism in carrion, or starvation because there is no carrion. Or by a bear because he cannot fit into a hole when his smaller sibling can. Or a snakebite because with his superior speed and strength he got to the snake first? Is all that so very predictable? And if not, how do you make it out to be less random than genetic change? And if it is, how do you correlate it with genetic change?
Let’s change the subject and move on to sentence 2.
>...however genetic variation in single organisms is not evolution. <
***Really?*** We learn something new every day! Firstly, I assume that what you mean is that genetic variation between the organism and its parent(s) is not evolution? (If not, please enlarge; you would have me completely nonplussed!)
Frankly, I cannot treat that assertion seriously. Firstly, there is for a start a difference between generations, which by definition is evolution. OK, I grant that statistically speaking such a tiny change is typically undetectable or apparently negligible, but remember that a typical mutation begins with a single event in a single organism. Natural selection does not say non-randomly: hey, time for a selection event; lets have a new gene appear in one thousand flies (or flowers or mould cells, or sardines). A single nucleic acid change in a single cell is the start. It might take several subsequent changes to fix and optimise the effect in response to selection, but each of those in turn starts in one organism. Adaptive evolution is not a single event, a population suddenly switching from one phenotype to another; it begins by selection of one parent and develops from there.
Of course, if the original change is say, recessive or otherwise silent, growing adventitiously in the population, the frequency might (ahem!) randomly increase in the population because of founder effects and the like, but how are you going to argue that that is non-random?

>So that means based on genetic variation we can not conclude evolution is random. (Because most of these genetic variations are discarded)<
Sorry zomgwtf; that is a simple non-sequitur, even in the light of your own claims. Would you care to rephrase it?
>Evolution is partly based on the genetic variations so randomness does play a role but evolution is not dependent on random acts. That's to say, it doesn't matter what comes out of the genetic variations evolution will still happen.<
That was not a great deal better. Darwinistic adaptation depends on natural selection, ie differential reproduction in terms of heritable differences within a population. Theoretically this does not demand randomness in either genetic change or ecological circumstances. In practice however, there is little possibility to predict either the genetic changes, or the details of the empirical events.
So far so random!
***BUT*** given certain information about changes in the environment and the populations within it, we can make some kinds of predictions. Some would be of a general handwaving type, such as “No miracles please; this is science”. I would undertake to breed pigs with wings by simple selection (no GMOs for instance), given a few million years, but to produce flying pigs by natural selection in a human lifetime would not work, even though most of their genome is simply convertible to bat sequences. Conversely, we know very well that a typical generalist ground squirrel population living in a region between forest and savannah is almost bound to produce the likes of both tree squirrels and groundhogs or prairie dogs unless there are already competitors blocking some of those options.
That is where the predictability, the non-randomness of natural selection comes in. I think it was G.G. Simpson who coined the term “evolutionary opportunism”. When we know the selective pressures and the genetic resources, we can make a very good guess at the trends of adaptation.

>No where in theories of evolution will you find that evolution specifically chooses larger, faster, stronger, brighter etc. animals over another.<
Do I smell straw? I am not aware that I suggested that you would find anything of the type. However, I do propose that natural selection would indeed favour such creatures very specifically indeed, as long as a) there was nothing stopping it; and b) these attributes were favourable to their successful reproduction. In fact, you do not make it clear why you urge this, because it is exactly a point that illustrates the fundamental nature of non-randomness in evolution!
>It all comes down to inheritance of genes, the animals that pass on their genes carry on the evolution of their species. Period. <
Yes? I find it hard to take that with a straight face. Would you care to explain either what anyone said to deny it, or what it has to do with either the randomness or non-randomness of evolution?

>Nothing is random about this at all and over time it's clear that the animals that pass on their genes are the best adapted for the enviroment and have a higher probability of being chosen to mate with. <
Zomgwtf, that sounds sooo cogent! Nothing is random about it hm? I have never seen a finer example of proof by re-assertion. I have at tedious length (let’s see you deny that!) demonstrated aspects of randomness and non-randomness in all those aspects. It is simply the way the world works. Nothing random? Next time you see an oyster or salmon or cnidarian spawning, please predict which of the offspring will be “best adapted for the enviroment and have a higher probability of being chosen to mate with”. It is no good claiming *after* the event that you could tell which was which; if that was good enough, you would win every bet on the horses or the lotteries. Remember what Bohr said about prediction?
Sorry mate! That is what randomness does. If it weren’t random you *could* predict it. You might argue that it isn’t randomness that causes a fair coin to fall unpredictably, but then what would *you* call it? Bad luck?
Jon
sylas
#42
Apr3-10, 01:37 PM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by zomgwtf View Post
The genetic changes that occur are definitely random yes... no one has said otherwise. Evolution however is not individual genetic changes in an organism but genetic change in an entire population of species. It's not by chance that the organism gets to mate and continue the inheritance, it is something that occurs naturally for various reason.

Only in hindsight can we look back and say 'well it's lucky that humans evolved larger brains when they could have just as easily stayed with more primative brains,' the thing is though that it was not by chance but by natural selection.
It certainly IS by chance that an organism manages to mate and pass on their genes. Most organisms have a heck of a lot more than two offspring per generation, ten or twelve or twenty or thousands. In the long run, (for the simple sexually producing case anyway) two offspring manage it.

Or consider plants. Evolution works for all life and making sure the theory works across the board means thinking of them as well. I remain quietly confident that chance is absolutely critical for success in mating.

In fact, I would think for the great majority of organisms, it's not even close. When you have thousands of spoors or seeds going out, some with one mutation and some with another, what's the biggest factor? I propose it is chance.

One problem is that people mix up chance with the idea of 50/50. (Casinos love those people.) They aren't the same thing at all.

The effects of selection are to bias the odds, usually just a little bit. And that is enough to mean that, over long time spans, fitness of the population is usually maintained. Because when you have long time spans with thousands of generations, a little bit of selection can have a dramatic effect.

You seem to be speaking of natural selection as something deterministic. It isn't. In population genetics, a gene (or more correctly, an allele) is said to increase reproductive fitness if organisms with that mutation are more likely to have surviving offspring. But it is certainly not a guarantee.

There is a finite non-negligible probability for any allele to be either fixed, or eliminated in a population. Either result is evolution, by definition, because it is a change in the distributions of alleles within the population. I know it is not about individuals. That's why my earlier post consistently refers to populations throughout, and refers you to texts on populations genetics, which go into the maths involved.

If you have a stable population of N individuals (a nice simple ideal case to start explaining this things), then a neutral mutation has a 1/N chance of being fixed, and a (N-1)/N chance of being eliminated. There's no selection involved here at all, but it is still evolution, of course; by the definition we have both been using. (1/2N if we get into the whole haploid diploid thing.)

But what if there is selection involved. In that case a beneficial allele is one with a greater than 1/N chance of being passed on, and a detrimental allele is one with a less than 1/N chance of being passed on. In either case, the odds are stacked against being passed on... even for beneficial alleles.

The best you can say is that in the long run, the casino wins. But it does so with a random walk, and precisely where you end up depends on chance, to a considerable degree. So sure, I continue to think of evolution as "random", meaning not that all results are equally likely, but only that the outcome depends on chance.

You seem to suggest that it was inevitable that we'd end up with large brains. Why would you think that? Personally, I think luck or chance had a heck of a lot to do with it. The vast majority of other living organisms on this planet didn't go that road, so why us? You know what I think? Chance. And why not? Was it inevitable that the robust Australopithenes are now extinct, but the gracile Australopithecenes live on, indirectly, as their descendents in the Homo genus? I don't think so; but then I don't think natural selection is a deterministic forces that ensures a particular result with regard for chance.

I think if we were able to somehow wind back the clock ten million years and run it again, there's no assurance at all that large brained hominids would reach plague proportions in the present. They might, they might not. They did in the only time we've run the experiment, but to say this wasn't luck is kind of assuming the conclusion, isn't it?

Natural selection itself is not chance, I agree. But evolution is not natural selection. It is change in distributions of heritable characteristics of populations. Natural selection has an impact on evolution because it modifies the chances of life and love for individuals with different characteristics.

Cheers -- sylas

PS. Missed two posts while I was writing. I think I'll read for a bit before writing again. Pleased to meet you, Jon.
zomgwtf
#43
Apr3-10, 02:04 PM
P: 500
The evolution of larger brains WAS inevitable, given the conditions for survival, when they began to develop. Just because it was by chance that the brain somehow did change does not mean that it being inherited and passed on generation to generation is random as well.
sylas
#44
Apr3-10, 02:17 PM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by zomgwtf View Post
The evolution of larger brains WAS inevitable, given the conditions for survival, when they began to develop. Just because it was by chance that the brain somehow did change does not mean that it being inherited and passed on generation to generation is random as well.
Well, I have yet to see any actual argument for this inevitability; only the assertion. You might be right (but I don't think so). It doesn't follow from conventional biological evolution, or any modern biological theory. It sounds more like some kind of throwback to pre-evolutionary ideas.

Cheers -- sylas
Jon Richfield
#45
Apr3-10, 02:21 PM
P: 258
zomgwtf:

>Random- governed by or depending on chance;<
Taken out of context this is meaningless; tautological at best. In probability theory you need to do a LOT more than that. Also in simple good sense; this is simply the kind of informal factoid that lexicographers come up with when they have asked an expert for a definitive answer and have failed to understand that they have missed half the point.


>Chance- luck: an unknown and unpredictable phenomenon that causes an event to result one way rather than another<

<snggr!!!> zomgwtf, REALLY! Would you care to paraphrase that??? So chance - luck is a phenomenon that causes things huh? Like gravity?
Yeah... riiiiiightt!!! ROFL^n!


>Evolution- (biology) the sequence of events involved in the evolutionary development of a species or taxonomic group of organisms<

Try writing that in answer to an exam in evolution 101 and you would get a flat zero! Don't you know a tautology when you see it? Even apart from the tautological aspects, that is nonsense as it stands. This student must try harder!

OR

>change in the inherited traits of a population of organisms through successive generations<

This is a lot better, though not formally cogent. It hints at the right idea though. Not that it clarifies the concept of natural selection in relation to evolution, which is not at all the same thing, though the relationship between the two is commonly misunderstood. For people who don't know the context that could be badly misleading.

>Both meaning the same thing.<
zomgwtf <Haaawwwww-haaaah-haaa! da capo> That has to be the funniest thing you have said in this exchange! Have mercy, man!

>Of course randomness has effects on Evolution but that is far from being the same as saying Evolution is random. Period.<

Aaaaah! The soothing voice of sanity and authority. At last! All settled. Mmmm... yes. Sort of...

Errr. zomgwtf...

zomgwtf, ol' pal ol' chum, one little detail. Just who said that "Evolution is random"? Not me for sure. Not anyone else that I took seriously.
And: "Of course randomness has effects on Evolution"?
What is that supposed to mean? If I wrote that in an exam, I had better hope it was not an exam in evolution, or even in advanced general biology. I would be lucky to fail without a black eye from the lecturer for cheek!

>Period<?
Oh. Well at least that sounds authoritative!

go well

Exclamation mark!

And Sylas, many thanks for the welcome. Pleased to meet you too :-)

All the best,

Jon
sylas
#46
Apr3-10, 03:04 PM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by Jon Richfield View Post
And Sylas, many thanks for the welcome. Pleased to meet you too :-)
It is just possible that we've online before. I initially learned about evolution by reading and then cutting my teeth on discussions, on USENET. P*t*r Ny*k*s is a mutual, uh, acquaintance. My own name is Chris Ho-Stuart.

zomgwtf, some of this discussion will turn somewhat on the meanings of words. Other more definite statements (like a confident claim about a particular evolutionary outcome in a given environment) are a bit more than this.

However, I have a good friend, John Wilkins, who is a genuine academic Philosopher of Biology. (That's like philosophy of science, but a specific to biological science.) And we have a mutual net-friend (whom I have cited previously) in Professor Larry Moran, who is an evolutionary biologist paying never no mind to uppity philosophers.

When the two of them start writing papers at each other, the rest of us may not get a definite answer to the original question, but we can sure learn a heck of a lot about evolution!

Larry can be summarized as "evolution by accident" is an accurate description of how evolution occurs, pointing to how chance and contingency are deeply involved in every step of the way from the sources of new variations to the events which lead them to be passed on, or not; much as I have done. John, on the other hand, can be summarized as "evolution is not fundamentally a random process." He doesn't deny the obvious role of chance in mating, in survival, in mutations, living long enough to breed and so on. But he tends to draw a sharper boundary between the chances and the processes of evolution. Both are quite prolific contributors to the old talkorigins archive, and they continue to comment on each other's blogs on this and other matters.

I'll see if I can get some references, which shouldn't be taken as closing the debate, but seeing how various other people have approached this question.

Cheers -- sylas
tkjtkj
#47
Apr3-10, 05:03 PM
P: 41
Quote Quote by sganesh88 View Post
Sorry i was viewing the 2nd page again and again; didnt notice your post.

We can't precisely say how an individual subatomic particle will behave. Randomness seems to be the inherent nature of universe. Not just during the big bang. Even now. Forever presumably.
But can you exactly pinpoint where i have gone wrong in my argument (the X-Y scenario and the analogy derived for humans)? so i can correct myself. Thanks.
I try my best, but for some reason i'm unable to get thru .. I've demonstrated why randomness is *IMPOSSIBLE* as the sole determinant of evolution.. And yet, you still ask me to show , again, where you are incorrect.

I'm sorry if im unable to make myself clear. Once again, i can suggest you read the following text:

"Inside the Human Genome" by John C. Avise

..tho i feel that even after you've read his extremely clear, erudite discussion, you'll come back with the same questions.

tkjk, md <ret>
Jon Richfield
#48
Apr4-10, 03:47 AM
P: 258
tkjtkj,
I cannot agree where you say "One can not have 'constrained randomness'. " I would go further and ask where we can have unconstrained randomness. I have never seen an example.

Just to forfend any misunderstandings, I think that Darwinism is a class of process, the course of which is affected by innumerable "random" effects and events, but which none the less is far from random in its nature and effects, so let us not get side tracked. What I am speaking about now is about randomness, not natural selection.

As I have said elsewhere, selection (not necessarily in the Darwinian sense!) from an unconstrained set is hard to imagine. In fact, equiprobable selection from an infinite set, constrained or not, is a very slippery concept, fraught with paradox. No one in human history has managed to make a random selection from even the numbers in the interval 0-1.

No one ever will.

How do I know? Because there is not enough matter in the observable universe to represent any number from any *subset* with a non-zero probability of fair selection.

Right?

Randomness from our point of view has to do with lack of information. The information might not be there at all, if various physical theories have any substance (personally I conditionally accept those theories, simply because I am not equipped to deny them and they seem to work in practice, so my inability to see how things can "just happen" is irrelevant. Mind you, it does seem to me to have a lot to do with the Aspect experiments and the Bell inequality.)

However, one need not appeal to quantum theory. We live in a chaotic universe. Consider a thought experiment : you have two (perfect) pool balls on a (perfect) pool table. One ball is at an arbitrary point on the surface, such that it is not perpendicular to any wall where there is a pocket. Put the cue ball on the surface perpendicularly between that ball and the table wall.

Got it so far?

Now Mr Super-player gives the cue ball a hefty whack so that it hits the first ball against the wall, and bounces against the cue ball. The two balls merrily go bouncing back and forth, and being perfect balls on a perfect table, they go on bouncing on the same line forever, right?

We had better assume that Mr Super-player is very symmetrical of course, because think what would happen if the balls were to strike each other off-centre by even the minimal Planck's distance, what... 10^-40 m? how many impacts would it take before we got a complete miss? If you bother to calculate it, you will be shocked at how quickly it would happen. In real life, you would be doing very well to get even *two* collisions (try it if you don't believe me!) Even the gravitational effect of Schroedinger's cat sneezing in its box 10000km away round the planet would disturb our ideal balls more than enough to prevent a long series of impacts.

Now, randomness in evolution has little to do with equiprobability, but it has much to do with unpredictability, and precise predictability is alien to our universe. To understand the role of randomness in evolution one has to understand those facts even more than one has to understand the nature(s) of randomness.

Just a few thoughts...

Jon

PS Oh, and btw, if anyone still thinks that evolution is random in any naive sense, let me refer him once more to the list of books I mentioned earlier in this thread. They are by far not the only ones, and any aficionado will have his own list, but they should suffice, no matter how one defines randomness in any defensible way.

PPS. Let's stop again for a think, thank, thunk. Let's reject all hidden variables in physics, sticking to the purest QM view of reality. OK? Now, that means that the basis of any event or effect in nature is fundamentally random, right? OK again?
That implies that all events in nature, whether the snozing of a sneeze, the falling of a stone, the running of a race, the raising of a skyscraper, the digestion of a steak, or the absorption of a microgram of Po, is random, right?

Rrrr... errr... well...

Maybe not quite so right. In another thread they are discussing the addition of random numbers and proof that the results must also be random. You might wish to check on that discussion. Here I merely remark that there is food for thought.
J.
Jon Richfield
#49
Apr4-10, 04:02 AM
P: 258
Sylas,
>It is just possible that we've online before. ... P*t*r Ny*k*s is a mutual, uh, acquaintance. <

Oh, right! I had almost forgotten that!
Such contacts, whether on USENET or any constructive forum are in my opinion enormously educational if one takes them in a suitable spirit. The gentleman in question seems to have taken our debate less affably, and did not respond to my greeting the next time we encountered each other.

Pity, still... plenty of other congenial spirits on line! For example, in a months correspondence I cannot remember his making a single substantial statement, whereas in this present forum I disagree with much that some people say, but at least they are saying things and mostly staying friendly!

My respects and thanks accordingly, to all of you!

All the best,

Jon
sganesh88
#50
Apr4-10, 08:17 AM
P: 288
Jon and Sylas, can i have your thoughts on my posts? especially this post no.27
http://www.physicsforums.com/showpos...3&postcount=27
sylas
#51
Apr4-10, 10:19 AM
Sci Advisor
sylas's Avatar
P: 1,750
I had been thinking of doing this anyway, sganesh88, so here goes.

Quote Quote by sganesh88 View Post
I think i need to specify clearly what i mean by "random" and how i conclude evolution is, according to that definition, random. Ofcourse when a bear falls from a mountain, its initial downward acceleration is 9.8m/s/s.-> not random. In the group of fruitflies you mentioned, those with wings survive(no fan scenario)-> not random.
I don't quite get where you are going here. You provide examples, but not a definition.

But look at the origin of all these. This touches upon the concept of freewill and let me be clear that i DON'T support creationism(I hope that will make you read the whole post without sarcasm)

Can an organism by itself "decide" the outcome of events and work towards it? . When an organism X confronts a situation (Like the casino situation you mentioned), its history helps it in dealing with it. Courtesy:Genes. If i am a good mathematician, I can easily calculate mentally and have a better chance of winning the game--> Not random in a superficial way. But what made me a good mathematician? Was that consciously decided by me independent of my past history? or was the decision to pursue math deterministic?
I don't know why we are talking about free will here.

Let's go to a hypothetical scenario. Two groups-> Predator X and prey Y. In the Prey group Y, all the members individually get affected by the predator group X. Some random mutation occurs in one particular individual of the Y group leading to the formation of rudimentary eye that helped it escape from X. It then went far, mated with other Ys. A group of Ys with eyes became commonplace after several generations because other Ys were eaten away by Xs.
This is what we call Natural selection. But think about other unfortunate individuals of the Y group. Does this mean they have simply no role in the encounter with X group? They couldn't have done anything. They went down without any conscious fight. Meaning that their fate isn't in their "hands".
I get the point, but the particular example is dubious. You can't form an eye from a random mutation. The formation of an eye requires a whole pile of mutations, and time for them to arise and be fixed in a population. It's a favourite example for creationists, who mistakenly think there can be no selective advantage to these mutations until you have an eye in place. But let's not side track into evolution of the eye, which is well discussed in a number of places.

A more plausible example would be mutations which made the prey species a little bit closer in color to the sandy sea floor on which they live, or a little bit more different. When a predator moves in to grab a meal they are a bit more likely to end up eating a prey individual that is seen more easily. Over time, the mutations that become fixed in the population are more likely to be those that help make the prey closer in color to the sea floor.

Under these conditions, the tendency to the coloration might seem inevitable. That's not necessarily the case, however. There are other strategies that might have been hit upon; such as developing a propensity to feed amongst the darker seaweed, or burrow in the sand.

About the only certainty is that there will be evolution.


Analogous to this, Evolution says that the determination of greats like Gandhi,Einstein and Abe Lincoln are just random(random according to this definition) because they were just groups of cells acting in unison and the genes instructing them to behave in such a way in such a situation(not random in a superficial way). There was never a conscious fight. If there seemed to be one, it was just an illusion. Am i right in understanding Evolution in this perspective?
It is not even clear that the determination of "greats" has anything particularly genetic, nor that it gave any reproductive advantage. I don't think evolution says much about this, and we've got the whole question of nurture vs nature showing up.

Ghandi had four children, and a fair number of grandchildren; but not exceptionally so. He was in many ways a lousy parent.

Einstein had three children, one of whom gave him grandchildren.

Lincoln has no surviving descendants at all; the last died in 1985.

It seems to me that you have mixed up a personal and humanist view of greatness with a biologial and evolutionary view of fitness.

-----

There's another thing I have been meaning to add into this thread. It's a common presumption that evolution leads to "better" and "better" organisms, in some sense. This can be misleading.

Living creatures are already well suited to their normal environment. The major competition they have is with others in their own species.

Frequently, the changing characteristics within a population is a crucial part of the environment for determining fitness. An animal which lives quite nicely thank you at one time might not fare so well if suddenly transported to live with its own descendants a hundred thousand years hence, even if the environment is otherwise identical. It would likely be outcompeted. But, ironically, the individual a hundred thousand years hence suddenly restored to its ancestors might also fare less well, in that it has been moved out of the milieu for which it was adapted.

A very common feature of evolutionary change is related to sexual selection; natural selection in the way of winning a mate. To this end, many creature evolve bizarre structures and forms. Impossibly overweight antlers. Flashy plumage or enormous gaudy tails. A bright red rump. And so on. And perhaps also a large brain fits in here as well. The idea is seriously proposed and there's no reason to reject it on evolutionary grounds. It's a hypothesis to be tested, and it can be tested indirectly; correlating sperm competition with brain size in primates.

Precisely what features end up being selected in this way can depend on accidents of mating preference and this MIGHT actually be something "chosen" for a creature with the capacity to think and reflect on the matter, if we acknowledge such a thing as free will. That's partly why I think the free will thing is a distraction that misses the real points at issue here.

Cheers -- sylas
zomgwtf
#52
Apr4-10, 10:47 AM
P: 500
Ah sylas, when I had mentioned it would be 'ineivitable' I meant only under the specific conditions which were available. Including the fact that the ability to have larger brains was still present.

This is the same thing for you color changing prey. The ability for the prey to be a closer colour to the waters floor combined with the fact that this will give it a greater probability to live to reproduce will make it ineivitable that this colour will become more prevelant in this organism. OR that it will further evolve it's colour in subsequent generations but the original change in colour will still have occured.

I agree with everything you have written thus far except that you consider evolution random. Evolution isn't random (by chance) it is something that just occurs naturally.

The way that I see this situation is that if we went back to a specific evolutionary event, say one allele dominating in the population. If we went back to the time period which this allele was existing and the conditions were exactly the same then it would again become dominant in the population, given that all conditions were the same. If it were random or based on chance however, it may turn out the same way or it may occur differently. Depending on what chance it has for either situation, this is not evolution.

I do not think that fact that I have evolved say a smaller jaw size is in any way based on chance. Even though it could have occured differently the prevalent conditions at the time allowed for it to occur. Even if the conditions are random that doesn't make evolution itself random either.

Am I wrong in my thinking here? This is what I have learnt of evolution in the past few years. From various sources, including talkorigins which you mentioned sylas.
Jon Richfield
#53
Apr4-10, 02:35 PM
P: 258
Sganesh,

Most of what you say is reasonable in various lights, but not much of it can be answered with simple yeses and noes. Let’s have a go at some of it.

Firstly, what is randomness and what does it have to do with Darwinism? That one has been chewed loudly to rags, both here and elsewhere. Views have ranged from variations on “Nothing is random” to “Everything is random”, with a lot of variations in between. Unfortunately, being in between is no guarantee of being right or even helpful. In this matter relevance is crucial. Although I have been repeating specifically that randomness is a function of how hard it is (how much information is required) to describe an effect, or alternatively, how unconfidently one can predict an outcome, one seldom can justify day-to-day evolutionary events on such principles.

Another problem is to put a finger on what constitutes a distinguishable evolutionary event. I have referred to “silent” genetic changes, such as recessive mutations that might not show up for generations after the first genetic accident <ahem!> that created the first of the line. Someone else (correctly) pointed out that many such mutations simply died out. Now, most such mutations are very hard to characterise; we used to think automatically of a genetic change as a DNA base change, or insertion, or deletion causing a change in the protein that a gene encoded. The more we study the matter however, the more options we find for how actively a protein or control string of nucleic acid is produced, and in combination with what other products and under what circumstances, and more things than I can think of offhand.

Then again, suppose we do in fact have a genetic change, what will its effect be? Physiologically important? Visibly phenotypically important? Speciationally important? Those things are not necessarily the same, please note.

Some recent work has strongly suggested that most speciation events are based on single, highly random genetic events, which makes it all look very simple, except that all such a speciation event usually would do is to render two sub-populations effectively non-interbreeding (reproductively isolated, as they say). The daughter populations still have to undergo the adaptive selection that will establish them respectively as viable independent populations, by processes that need not at all be equally arbitrary and abrupt. And those processes need not be any less “random” in their advent, though they could be far from random in their effect on the evolution (in any sense) of the population. And of course, if the populations compete, one might well drive the other to extinction. As a rule, if both populations survive it is because they have diverged into different ecological niches or parapatric distributions.

Now, evolution aside, and as a matter of logical principle, how can randomly adventitious events, as a general principle, lead to highly non-random processes? Surely and obviously, anything of the kind must be about as likely as closed-system decreases in entropy? As monkeys typing Hamlet?

Not so, but far otherwise! Let’s consider examples, hopefully illustrative.

Suppose I were to persuade you to invest in a lot of assorted tablets of water paints. I then talk you into applying water and a brush at random and stirring things up. What do we get? You know very well: maximised entropy, embodied in a muddy brown, indistinguishable, non-representational mess. To be sure, the result is not in all ways random; with remarkable consistency we always get much the same colour, non-random though such a consistent result may seem. But Mona Lisa shows there none, and nothing like any thrilling picture. Yes?

Well, next I talk you into buying a cement mixer (Using your money, I am too stingy for either experiment!) You set it up and I charge it with a barrow-load of stones and perhaps some water. I ask you to predict what we will get apart from a lot of noise. “Some grit I suppose,” you reply.

And you are right as usual. Mind you, when we stop the motor after a few hours of the neighbours complaining about the racket, we notice that the jagged stones, though still nothing for Henry Moore to brag of, do seem to have become less jagged... Now, how could that be? We could hardly have assaulted their peripheries more randomly, could we?

Now, let’s abandon these tedious scientific speculations and experimentations in favour of some sparkling reminiscences from my own life, exactly the sort of thing you have been hoping for.

When I was a little boy (my parents were honest but poor and all that sort of thing) I once spent a summer holiday at a not-yet-popular coastal village called Gansbaai. (You know Gansbaai? No? Oh…) Well anyway, it had a lot of sand and sandstone shingle. It still has them.

Except that in those days it among the rest of its shingle, it used to have a LOT of white sandstone rocks, remarkably smooth and strikingly precisely ellipsoidal (each of the mutually perpendicular planes of symmetry representing a different ellipse!) Nowadays the place is far more popular and you no longer find such stones without weeks of search. All wasted in the souvenir boxes of trippers.

Wasn’t that entertaining? Didn’t it bring sentimental tears to your eyes? No? Oh...

Well, then try this one:

In the local, largely sandstone-topped mountains, many streams, some seasonal, others perennial, run down and gouge great gorges out of the rock. There is a high incidence of minor waterfalls, and commonly there are potholes under the falls. The potholes are as a rule strikingly round and surprisingly deep, and many of them contained (mostly stolen by now) stone marbles typically some ten to twenty cm in diameter, and *beautifully* spherical.

“Now justasec JR!” I hear you cry, “what kind of fool do you take me for? You planted those rocks there, right? Bias is all very well, but perfect, smooth ellipsoids? And in special cases, spheres in cylindrical, smooth-sided holes? Pull the other one!”

Nope. I insist. Here we have an instance of a *constrained random process*. Such processes are of the most fundamental importance in all sorts of situations in nature. My pet rocks are just one class of myriads. *They are exposed to a pretty well random influence*, but by their very nature only their external surfaces are exposed to abrasion and spallation. What is more, their most distal and gracile parts are most violently assaulted. It is in fact a particular class of natural selection. (The fundamental way in which it differs from Darwinian selection, I leave as an exercise for the advanced student.)

Notice that although there are all sorts of unpredictable (“random”, right?) *aspects* to the process, the upshot is highly, though by no means perfectly *predictable*, very far indeed from substantially random. No paintpot-brown here!

Hmmm... unless you count paintpot-brown as a "strange attractor" logically analogous to such trends as for abraded stones to become rounded. Tricky things, these constrained random processes!

OK, have I laboured the point enough? Constrained randomness is perfectly possible, even omnipresent, and can lead to drastically non-random results without the slightest hint of intelligent design or conscious intention. It does not follow that intelligent agents cannot employ such principles or influence them; in suitable contexts that is called “technology”. For example, tumbled stones are common in cheap jewellery, right? Fundamentally the same principle as in the potholes.

Now, how far must I draw analogies between the constrained selection of the stones in tumbled abrasion and of strains of living creatures in adaptive selection? Only certain, suitable, stones gave spectacular results under certain circumstances. Only certain, suitable, genotypes, via their phenotypes yield high levels of viable adaptations under suitable, largely destructive, selection. Swap the selection for two genotypes, and you might get different outcomes, even extinction.

Deeply varied types and degrees of randomness give drastically different or eerily similar evolutionary upshots. I visited Australia some years ago , and I was repeatedly stunned by the degree to which just a few genera of plants produced large ranges of functional adaptations, and conversely, how practically unrelated genera produced practically indistinguishable results, not always independently! (Figure it out!) If you happen to go there, (especially the south and west coastal areas and outbacks, pay especial attention to Grevillia, Acacia, Banksia, Hakea, and a few more, particularly in their knee-high, arid-scrub incarnations. Amazing!) In South Africa there are many more examples, but because of the greater biodiversity, the effect is less striking, though it frequently makes a real fool of one! Especially the xerophytes.

Again, what one might call the micro-events in the evolutionary process are as random as one could make them, and teleology is not even on the horizon, but though the course of adaptation is not generally predictable except in minor degrees and contexts (Remember Orgel’s “second” law) there is no way one could rationally look at the course or outcome of the process and call it random.

Now, various people spoke of the mental processes or intentions (or lack thereof) in the participants in these processes. However, generally there was nothing like consciousness involved of course, or when there was, there was nothing like teleology. Not in natural selection anyway.

As soon as teleological artificial selection takes a hand, adaptation, though short-term, speeds up by orders of magnitude. We have bred no talking dogs or flying pigs, but what we do have has come about practically overnight, in thousands or even tens of years instead of hundreds of thousands or millions. (actually, there are counter examples, but never mind those for now. They are generally instances of concentrated selection focussed on particular adaptive requirements. )

More interestingly, we ourselves as a species are on the brink of an opportunity that has not occurred on this planet before. If only we could get our fingers out and our brains and goodwill into gear, we could be the first species in this neck of the galaxy to apply planned, conscious adaptation to its own evolution.

Try *that* one on for shivers on a hot day.

OK, I know that is not what you asked, but was it helpful?

Go well,



Jon
zomgwtf
#54
Apr4-10, 07:05 PM
P: 500
Jon I must say reading what you wrote was quite intriguing. I think you and I have pretty much the same view of Evolution and I'm not so sure why I felt that it was any different before.

One thing I like to add though is that regardless of what we see as the evolutionary outcome of any event does not effect how random the act of evolution is. As you illustrate it could occur any other number of ways which us humans in most cases would not be able to predict. Evolution would still occur though and it would not be random, it is something that just happens and generally speaking over an entire population over time it leads to better adaption to the current situation.


Register to reply

Related Discussions
Ecosystematic Evolution (symbiosis in the importance of evolution) Biology 5
Proof of the statement: sum of two random variables is also a random variable Set Theory, Logic, Probability, Statistics 8
Evolution not completely random according to an experiment Biology 3
Expectation and variance of a random number of random variables Calculus & Beyond Homework 3