Chance (and regularity) in evolution

[T'ai Chi]

Looks like another piece on randomness being involved in evolution:

http://scienceblogs.com/pharyngula/2006/03/chance_and_regularity_in_the_d.php

[NotJesus]

Looks like another pointless T'ai Chi thread.

[CapelDodger]

Looks like another piece on randomness being involved in evolution

"Looks like" is one thing, and terribly subjective; do you think it is?

[CapelDodger]

Looks like another pointless T'ai Chi thread.

And not even interesting :) .

[CapelDodger]

The cited article is very interesting, IMO.

[articulett]

The cited article is very interesting, IMO.

Yes--I'm a big pharyngula fan...
too bad Thai didn't understand it.

Although evolution takes eons, it eventually comes up with the best designs--because that which works gets pass on in a ratcheting process. Of course, humans can speed it up considerably as we do with race horses, dogs, and food crops-- An almighty sky dude would be able to do it instantly without the eons of slow and steady progress.

All the multi-trillions of replication experiments that didn't work--don't get to "exist" in the present. This confounds creationists to no end, it seems. I can't tell if woo makes you dumb or if the dumb or more likely to be drawn to woo.

[T'ai Chi]

Looks like another pointless T'ai Chi thread.

Another pointless comment. Why not try being a real skeptic and talk about the article?

Too much for you to do, I know that all too well. ;)

[Wowbagger]

For once, T'ai Chi actually links to an article that really is cool and interesting.
I just hope he pays attention to the parts that talk about the developmental process being "unguided", instead of misinterpreting it as "special design", like some are prone to.

[T'ai Chi]

I thought the randomness is quite interesting, since many drone on about how evolution is not random. :)

I can only assume they are ignorant of what evolution is comprised of, or are only knowledgable on the layman's definition of random.

[Zep]

When you talk about chance and regularity, TC, are you referring to roughage and bowel movements?

[RichardR]

Looks like another piece on randomness being involved in evolution:

http://scienceblogs.com/pharyngula/2006/03/chance_and_regularity_in_the_d.php

From the same author (http://scienceblogs.com/pharyngula/2007/04/the_sanctimonious_bombast_of_g.php):

Ah, and he dredges up the old creationist caricature of evolution, that it is a purely random process. Whenever someone tells you that it is, it's a flashing indicator that he doesn't know what he is talking about. Evolution is not random, although components of the process are.

Yes, that seems to clear things up.

[Walter Wayne]

I hate to say it, but it suggests to me that the author is somewhat ignorant of what random means.

Most people admit that winning the lotto 6/49 is random, and the odds of that are winning that are what ... zero ... to several significant figures.

Walt

[wollery]

I thought the randomness is quite interesting, since many drone on about how evolution is not random. :)

I can only assume they are ignorant of what evolution is comprised of, or are only knowledgable on the layman's definition of random.Find someone here who claims that the mechanisms of evolution aren't random.

Please, have a go.

[The Great Hairy One]

Read paragraph 3 again. Quote: "it's reliable, but not rigid".

PZ Myers knows what he is talking about when it comes to evolution. And, Walter Wayne, I can assure you he does know what he is talking about.

T'ai Chi, yes, evolution does contain random elements. But as RichardR posted, "Evolution is not random, although components of the process are."

Cheers,
TGHO

[athon]

I thought the randomness is quite interesting, since many drone on about how evolution is not random. :)

This strawman needs a hat and a carrot for a nose, I think. :rolleyes:

Randomness is indeed part of evolution. A bit like electricity helps a car start. Saying that cars are electrical is misleading, though. Just as saying evolution is therefore random.

But please, spin and twist away.

Athon

[fishbob]

Another pointless comment. Why not try being a real skeptic and talk about the article?

You first.

[JoeTheJuggler]

I thought the randomness is quite interesting, since many drone on about how evolution is not random. :)

I can only assume they are ignorant of what evolution is comprised of, or are only knowledgable on the layman's definition of random.

Tai Chi--do you think evolution by natural selection is random? (This is a yes or no question.)

[Jekyll]

Although evolution takes eons, it eventually comes up with the best designs--because that which works gets pass on in a ratcheting process.

As a minor nitpick, evolution is a form of gradient descent and so gets stuck in local minimas.

There is no guarantee that we'll end up with the best design for our situation, just a local optima. This explains why we still have blind spots, and a tendency for bad backs, it would require major structural reworking for us to be otherwise.

[Dancing David]

Another fascinating aspect of development is that all the intricate, precise steps are carried out without agency: everything is explained and explainable in terms of local, autonomous interactions. Genes are switched on in response to activation by proteins not conscious action, domains of expression are refined without an interfering hand nudging them along towards a defined goal. It's teleonomy, not teleology. We see gorgeously regular structures like the insect compound eye to the right arise out of a smear of cells, and there is no magic involved—it's wonderfully empowering. We don't throw up our hands and declare a miracle, but instead science gives us the tools to look deeper and work out (with much effort, admittedly) how seeming miracles occur.



....
I know, it's all a little bewildering and complicated—intensely complicated with all kinds of interactions between cells. However, when you dig into it and explore the literature, what you find is the successful application of a reductionist program of study, with each piece of the story a fully comprehensible and actually rather simple product of a molecular/cellular interaction. Complexity is the result of simple, repetitive, iterated processes which can yield regularity and chance variation…but at no point are there any events beyond local chemistry and cell biology.

[Dymanic]

Although evolution takes eons, it eventually comes up with the best designs--because that which works gets pass[ed] on in a ratcheting process.
Things are quite a bit messier than that. For one thing, "best" is a rather elusive concept. Does the male peacock's tail represent the "best" possible design? Is there such a thing as a "best" design for a car? We might begin by assuming that there is, and begin looking for it by looking at those designs which are most successful in the marketplace, but such an approach surely ignores a great many factors which may influence that success. We might decide to deliberately ignore those other factors, and simply define "best" as "most successful in the marketplace", but this doesn't seem like much of an improvement. You can rearrange stuff in your head like that all day without really accomplishing much.

As Jekyll points out:

"There is no guarantee that we'll end up with the best design for our situation, just a local optima."

A region of design-space may include numerous optima, but it's important to remember that they aren't all equally accessible from a given point. Due to the ratchet-like nature of the process,
once a population begins to approach one optimum, the others soon become inaccessible. Backing up doesn't work.

It gets worse when you consider the myriad ways in which "best" is subject to being constantly redefined by changes in environmental factors, population dynamics, etc. Is the "best" reproductive strategy to produce large numbers of weak, slow, stupid offspring, or is it to produce fewer numbers of strong, fast, smart ones? It depends -- on how plentiful resources are, on what strategy is most prevalent among the competition, and who knows what else.

All the multi-trillions of replication experiments that didn't work--don't get to "exist" in the present.
But a lot of the ones that did work don't exist in the present either.


at no point are there any events beyond local chemistry and cell biology.I have to disagree with that. At nearly every point, there are events WAY beyond local chemistry and cell biology, events which may significantly alter the course of evolution. There are tsunamis and wildfires and landslides and meteor strikes. A superbly adapted creature may be struck by lightning or tumble from a cliff, while its skinny flatfooted pregnant cousin is swept out to sea on a floating log and carried to a distant island where she becomes the founding member of a new population. At a level of high resolution, the stochiastic processes which drive development obey certain rules, and may be approached mathematically with some reliability -- but in addition to that, there will always be random macro-events which will defy all predictibility.

[slyjoe]

It always surprises me that a discussion of complexity theory never comes up when discussing randomness and evolution. As James Gleick noted in "Chaos", "...simple deterministic models could produce what looked like random behavior".(page 79). I think Jekyll hinted at the effects of this in post 18.

[Wowbagger]

I thought the randomness is quite interesting, since many drone on about how evolution is not random. :)When mutations occur, they are random with respect to whether they will be beneficial, detrimental, or neither, to the life form.

Selection is what is not random: Those mutations that are beneficial will likely survive better than those that are not, based on the pressures the current environment places on survival.

[aggle-rithm]

Looks like another piece on randomness being involved in evolution:

http://scienceblogs.com/pharyngula/2006/03/chance_and_regularity_in_the_d.php

Is this article supposed to support the idea that evolution is random, which therefore supports the idea that evolution is bogus because there is no way random events could result in such rich diversity and complex design?

Because the article, in supporting one small part of your premise, demolishes the rest. It explains precisely HOW random events can lead to complex design autonomously.

[Foster Zygote]

Is this article supposed to support the idea that evolution is random, which therefore supports the idea that evolution is bogus because there is no way random events could result in such rich diversity and complex design?

Because the article, in supporting one small part of your premise, demolishes the rest. It explains precisely HOW random events can lead to complex design autonomously.

You beat me to it. T'ai seems to think that evolution must be defined as either completely predictable or completely random. He seems to want to apply the term "random" to evolution in the same way that many creationists do: The "hurricane/747" argument. I don't know why he refuses to/is unable to acknowledge that a process can have random elements and still have a good degree of predictability and be highly structured.

[Walter Wayne]

I hate to say it, but it suggests to me that the author is somewhat ignorant of what random means.


PZ Myers knows what he is talking about when it comes to evolution. And, Walter Wayne, I can assure you he does know what he is talking about.
I wouldn't suggest he isn't an expert on evolution.

I don't understand why people keep insisting that evolution is not random. Is it reliable? For the most part. Do people here think it is reliable in small populations? Evolutionary lineages go through choke points, either due to a small number of individuals being isolated from the larger group or stresses on the population that result in generations with a small number of breeding individuals. It is at these points that chance and probability have the opportunity to produce a variance in the result. At the point where a mutation occurs is precisely when it is most vulnerable, as it is carried by few individuals, and in a small total population the odds of that mutation occuring again within a few generations is incredbily small.

Evolution is not like dice, with each throw being independent of the last. If I make a prediction on the results of a million rolls, even if the percentages look a little off after ten rolls I can be confident that my prediction will be pretty accurate after the scenario plays out. But in evolution, the preceding generation affects the next. The base of any evolutionary branch leaves its mark on the descendents. If that base is a small population those random elements will have a large affect on the result. Hence the randomness of evolution.

I think people see other random processes that might as well be deterministic (throw a billion probabilistic photons around and you get classical electro-magnetics), and assume you can say something similar about evolution, that it will wash out when the numbers roll in. But times when populations are small, as rare as such events may be, leave there mark on subsequent generations and so this random variation comes through.

Walt

[T'ai Chi]

Tai Chi--do you think evolution by natural selection is random? (This is a yes or no question.)

My thoughts are contained here

http://www.statisticool.com/evolution.htm

[Tief]

My thoughts are contained here

http://www.statisticool.com/evolution.htm

So, maybe you could explain how you came up with your formula,

Evolution=NS(RM, OS)

Seems to me that this is not mathematical in and of itself, but an assumption, and a huge and erroneous one at that.

I would ask, are you saying that evolution is random in the same way that, for example, coin tosses are random? If you were talking about genetic changes only, I might agree with you, though there are complexities involved in gene mutation that aren't involved in coin tossing. However, evolution is not defined by gene mutation alone, it is also defined by natural selection. To call evolution random is to ignore this important characteristic.

All that aside, I think the point is not whether or not you call evolution a random process, but whether or not you mean it in the same way as when you talk about coin tossing. Creationists do not make a distinction between the two. There is obviously and clearly a huge difference, and to ignore the difference is dishonest.

[articulett]

You beat me to it. T'ai seems to think that evolution must be defined as either completely predictable or completely random. He seems to want to apply the term "random" to evolution in the same way that many creationists do: The "hurricane/747" argument. I don't know why he refuses to/is unable to acknowledge that a process can have random elements and still have a good degree of predictability and be highly structured.

Indeed--because of the trillions of random experiments, we can definitely expect those that further the passing on of any particular stretch of DNA will become spectacularly widespread in a community depending upon the survival and reproductive advantageous it confers--and the stuff that doesn't work, won't survive to be observed.

The 747 argument makes much more sense if you think of a 747 evolving from the first airplane--in fact all airplanes can consider the first airplane a common ancestor--these are the designs that worked--fit the niche--were refined and honed via "artificial selection" by humans. Nature takes a lot longer to select, and there is a lot of waste and suffering in the process, but the designs fit the environment very well. Tornados are not eons long--and they tend to break things into small particles rather than assemble them into structures. The analogy is similar to thinking a flood could carve the grand canyon in an instant (ha). The monkeys typing shakespeare analogy is similar--except to be more similar to evolution, every once in a while when a word was formed accidentally it would stick on the page, while the nonsense just disappeared.

Besides, 5 million years after the split an ape descendant did type Shakespeare while his evolutionary cousins carved out their niches in jungles and on African Savannas. Creationists can't seem to understand, the power of time, natural selection, and the ratcheting process of evolution. I mean, they can understand it if applied to the internet...or to the creation of a mountain--but something in their egos makes them certain it cannot apply to
something as extraordinary as them.

200 million sperm per ejaculation meeting up with 1 in 50,000 eggs being ovulated, and still the results can be less than stellar. Creation is a bumbling process with occasional spectacular results. Thai is not one of them.

[CapelDodger]

As a minor nitpick, evolution is a form of gradient descent and so gets stuck in local minimas.

There is no guarantee that we'll end up with the best design for our situation, just a local optima. This explains why we still have blind spots, and a tendency for bad backs, it would require major structural reworking for us to be otherwise.

Quite. Evolution doesn't do back-tracking at all well. What evolution does is favour those who took the right track in the first place, purely by chance. Or took neither and dithered around as generalists.

[CapelDodger]

Creationists can't seem to understand, the power of time, natural selection, and the ratcheting process of evolution.

They can't because they don't want to. Only the present is real to them. Their arguments always work from here - soul-bearing god-shaped humanity as the culmination of existence itself - to how unlikely such an obviously desirable outcome is to emerge randomly. The idea that this outcome, which includes them and their short lives, might not have any cosmic meaning is not conceivable.

[JoeTheJuggler]

My thoughts are contained here

http://www.statisticool.com/evolution.htm


I knew you wouldn't answer my simple yes or no question with a "yes" or a "no". I read your link. I'd encourage you to take a course on evolutionary biology.

At any rate, the closest you come to answering my question is this: "It is true that natural selection is not random, but evolution is more than natural selection."

Right, there's also artificial selection, sexual selection and genetic drift (basically no selection).

If you think evolution is a random process, you're simply wrong.

I suspect you argue that it's random as a straw man so you can ask how such incredibly complex things could have evolved completely at random. The whole point of the theory of evolution is that it explains exactly that. It is not a random process.

[Foster Zygote]

I suspect you argue that it's random as a straw man so you can ask how such incredibly complex things could have evolved completely at random. The whole point of the theory of evolution is that it explains exactly that. It is not a random process.

That's exactly the impression I get as well. Even his writings on his website regarding evolution seem hostile, or at least pedantic.

Many sources make the case that evolution is not random, and to call it random is a horrible offense. Richard Dawkins, for example, wrote

The belief that Darwinian evolution is 'random' is not merely false. It is the exact opposite of the truth. Chance is a minor ingredient in the Darwinian recipe, but the most important ingredient is cumulative (natural) selection which is quintessentially nonrandom. (The Blind Watchmaker, p. 49)


Kenneth Miller wrote

Evolution is not a "random" process, and to characterize it so seriously misleads students. Natural selection, the most important force driving evolutionary change, is not random at all, but an observable, verifiable process that fine-tunes variation in populations of a species to the demands of the environment in which they live. It is true, of course, that variation in a species arises from sources such as mutation and sexual recombination, which are inherently unpredictable. Therefore evolution, like any historical process, can be influenced by random forces.

He presents these quotes out of context without explaining to readers that Dawkins and Miller are both addressing creationist claims that evolution is entirely random. Even in the quotes he's chosen to present both biologists acknowledge that the mechanism of evolution has a random element.

[Walter Wayne]

http://en.wikipedia.org/wiki/Stochastic_process
A stochastic process, or sometimes random process, is the counterpart of a deterministic process (or deterministic system) considered in probability theory. Instead of dealing only with one possible 'reality' of how the process might evolve under time (as it is the case for solutions of an ordinary differential equation, just as an example), in a random process there is some indeterminacy in its future evolution described by probability distributions. This means that even if the intial condition (or starting point) is known, there are more possibilities the process might go to, but some paths are more probable and others less.
Joe, do you disagree with the definition above, my bolded part in particular?

If not what part do you disagree with?

If yes how does that not apply to biological evolution?
(Aside: I state biological here to be explicit that I am not equivocating using the term evolution in the definition given, which uses it in the more general sense).

Walt

[JoeTheJuggler]

http://en.wikipedia.org/wiki/Stochastic_process

Joe, do you disagree with the definition above, my bolded part in particular?

If not what part do you disagree with?

If yes how does that not apply to biological evolution?
(Aside: I state biological here to be explicit that I am not equivocating using the term evolution in the definition given, which uses it in the more general sense).



I think "random" is being used here in a completely different way than what this thread is talking about. In the wiki quote you cite, "random" means strictly "not deterministic". No, no one is arguing that evolution is deterministic--that you can somehow predict which future forms will be successful based on the present. Evolution doesn't work that way, and there is NO ONE suggesting that it does.

I reject that the alternative to the deterministic view (again, a view which NO ONE espouses) is that evolution proceeds randomly. (Again, I think the wiki-quote is using a very narrow and uncommon definition of "random" as meaning simply "not deterministic".) I know for sure that forms like the elephant won't grow wings and develop flight.

Tai Chi and others are setting up a straw man where evolution happens by chance--complexity arising out of nothing like a wind blowing to assemble the parts of a 747. If you define evolution that way (and only evolution deniers are attempting to do so), it's easy to prove that complex structures aren't likely to arise. I've seen them even calculate "odds against" the development of the human eye, for example, by "random evolution".

Some kinds of variation (as in point mutations) happen randomly, but that's not the same at all as saying "evolution is random" or even Tai Chi's latest, that evolution "involves" or "is based on" randomness.

[Walter Wayne]

I think "random" is being used here in a completely different way than what this thread is talking about. In the wiki quote you cite, "random" means strictly "not deterministic". No, no one is arguing that evolution is deterministic--that you can somehow predict which future forms will be successful based on the present. Evolution doesn't work that way, and there is NO ONE suggesting that it does.

I reject that the alternative to the deterministic view (again, a view which NO ONE espouses) is that evolution proceeds randomly. (Again, I think the wiki-quote is using a very narrow and uncommon definition of "random" as meaning simply "not deterministic".) I know for sure that forms like the elephant won't grow wings and develop flight.

Tai Chi and others are setting up a straw man where evolution happens by chance--complexity arising out of nothing like a wind blowing to assemble the parts of a 747. If you define evolution that way (and only evolution deniers are attempting to do so), it's easy to prove that complex structures aren't likely to arise. I've seen them even calculate "odds against" the development of the human eye, for example, by "random evolution".

Some kinds of variation (as in point mutations) happen randomly, but that's not the same at all as saying "evolution is random" or even Tai Chi's latest, that evolution "involves" or "is based on" randomness.
Layman's definitions of random appear to me to fit just as well as the more rigorous definition.

An architect might build a structure with inconsistent (random) spacing of support columns.

A pinball machine is governed by basic Newtonian laws but, unless you know the position and speed of the ball precisely, it's position after hitting a few bumpers is unpredictable (random) because the errors build on eachother.

A crime is said to be random because the victim was a stranger chosen by happenstance.

Is evolution unpredictable? Yes.

Is evolution haphazard? It certainly doesn't get to the end products by the shortest possible route. Evolution didn't take the first lung fish to crawl onto land and think "so what is the shortest possible between this and a bipedal creature with a large cranium and opposible thumbs?" We got here by a route the know sane engineer would take.

What being chosen by happenstance? Well I can be selected and a friend deselected simply because it was his turn to be at the front of the hunting party when we found that quicksand trap. Or on a larger scale a lineage with potential might be cut short because it started in the meditarian basin around the time the atlantic ocean decided to make an entrance.

The ONLY definition of random I am aware of that doesn't fit is that of an event where all outcomes are equally likely but, as I pointed out elsewhere, that means that the sum of rolling two dice and adding them together isn't random because a 7 is more likely than a 6.

Walt

[JoeTheJuggler]

So are you saying there is no such thing as evolution by natural selection?

Selection means not random. That's really the whole point.

[CFLarsen]

Is evolution haphazard? It certainly doesn't get to the end products by the shortest possible route. Evolution didn't take the first lung fish to crawl onto land and think "so what is the shortest possible between this and a bipedal creature with a large cranium and opposible thumbs?" We got here by a route the know sane engineer would take.

*BING*

Evolution doesn't result in "end products".

[Taffer]

I wouldn't suggest he isn't an expert on evolution.

I don't understand why people keep insisting that evolution is not random. Is it reliable? For the most part. Do people here think it is reliable in small populations? Evolutionary lineages go through choke points, either due to a small number of individuals being isolated from the larger group or stresses on the population that result in generations with a small number of breeding individuals. It is at these points that chance and probability have the opportunity to produce a variance in the result. At the point where a mutation occurs is precisely when it is most vulnerable, as it is carried by few individuals, and in a small total population the odds of that mutation occuring again within a few generations is incredbily small.

Evolution is not like dice, with each throw being independent of the last. If I make a prediction on the results of a million rolls, even if the percentages look a little off after ten rolls I can be confident that my prediction will be pretty accurate after the scenario plays out. But in evolution, the preceding generation affects the next. The base of any evolutionary branch leaves its mark on the descendents. If that base is a small population those random elements will have a large affect on the result. Hence the randomness of evolution.

I think people see other random processes that might as well be deterministic (throw a billion probabilistic photons around and you get classical electro-magnetics), and assume you can say something similar about evolution, that it will wash out when the numbers roll in. But times when populations are small, as rare as such events may be, leave there mark on subsequent generations and so this random variation comes through.

Walt

There are a few processes in the theory of evolution which could be described as random. The chance of a specific locus having a Single Point Mutation can be determined to a set value but, much like flipping a coin, it might not happen. Mutations at individual loci is random, in that it cannot be predicted. However, the overall rate of mutation can be predicted extremely well, much like the overall number of "heads" and "tails" from a thousand coin flips can be extremely well predicted. Furthermore, there is often clear evidence that certain mutations are far more common then others. SPMs are not nearly as simple as "happens" or "doesn't happen". What's more, the mechanisms behind SPMs is fairly well understood. It might turn out to not be random at all. After all, in most cases it does not involve anything "quantum".

[Thabiguy]

Guys, guys! You are getting too worked up over a non-issue.

Bottom line is, it does not matter whether you call evolution "random" or "non-random". This is only a semantic question of the definition of your own terms. Such decision has absolutely no implications whatsoever regarding the validity of the theory of evolution itself.

It only becomes relevant in metaphysical arguments and distorted creationist lies (http://www.talkorigins.org/indexcc/CB/CB940.html), which I believe is the context in which it was mentioned by the quoted evolutionists.

Depending on the point of view, you may call evolution 100% random (seeing the universe as non-deterministic) or 100% non-random (seeing the universe as deterministic) and neither view changes the essence of evolution the least bit. It's just word play.

[T'ai Chi]

So, maybe you could explain how you came up with your formula,

Evolution=NS(RM, OS)

Seems to me that this is not mathematical in and of itself, but an assumption, and a huge and erroneous one at that.


It is a conception based off of what Dawkins, Miller, and many others, who say that evolution is the selection of mutations and other things.

[T'ai Chi]

I knew you wouldn't answer my simple yes or no question with a "yes" or a "no".


I'm not your lap dog and you don't get to tell others how to respond in discussions. And yes, you will actually have to read and exert some work if you really want to understand.


If you think evolution is a random process, you're simply wrong.


Wow, such detailed arguments! Look, then just show me a non-trivial process with any random inputs whose result you can predict with 100% certainty. It's there waiting for you to do.


I suspect you argue that it's random as a straw man so you can ask how such incredibly complex things could have evolved completely at random.


No clue where that came from.

[T'ai Chi]

Tai Chi and others are setting up a straw man where evolution happens by chance--complexity arising out of nothing like a wind blowing to assemble the parts of a 747.


Well that's clearly a lie to anyone that has read what I've written, and what others here have written.

It seems like the people who say things like 'well, evolution a stochastic process.... but it is not random!' are the ones denying reality, becase they have a strong emotional reaction to the word random, or a deep misunderstanding of its meaning.

[CFLarsen]

It is a conception based off of what Dawkins, Miller, and many others, who say that evolution is the selection of mutations and other things.

Please quote Dawkins, and explain how you got to this "technical" formula.

Wow, such detailed arguments! Look, then just show me a non-trivial process with any random inputs whose result you can predict with 100% certainty. It's there waiting for you to do.

Who has claimed that evolution can be predicted with 100% certainty?

No clue where that came from.

It's not a strawman to claim that evolutionists claim evolution can be predicted with 100% certainty?

[JoeTheJuggler]

I knew you wouldn't answer my simple yes or no question with a "yes" or a "no".

I'm not your lap dog and you don't get to tell others how to respond in discussions. And yes, you will actually have to read and exert some work if you really want to understand.
Then I suggest again that you take a course in evolutionary biology. Then you won't need to invent a formula as if we don't know anything at all about the mechanisms of evolution.





If you think evolution is a random process, you're simply wrong.
Wow, such detailed arguments! Look, then just show me a non-trivial process with any random inputs whose result you can predict with 100% certainty. It's there waiting for you to do.

One component of variation (point mutations) is wholly random. Evolution by natural selection is not. I'll use your argumentative technique by pasting in some links to pages that I'd like you to read:

http://www.talkorigins.org/faqs/chance/chance.html
http://evolution.berkeley.edu/evosite/misconceps/ICchance.shtml



I suspect you argue that it's random as a straw man so you can ask how such incredibly complex things could have evolved completely at random.
No clue where that came from.

As I said, it's a suspicion, so it came from my mind. It's based on the track record of ID/Creationist arguments. They note examples of complexity; point out that you can't get complexity out of random action (I've even heard them cite the Second Law of Thermodynamics!), therefore an Intelligent Designer is the correct explanation. Do you deny that this argument has been made?

Are you a proponent of Intelligent Design?

[CFLarsen]

Are you a proponent of Intelligent Design?

Yes, he is.

[articulett]

As a minor nitpick, evolution is a form of gradient descent and so gets stuck in local minimas.

There is no guarantee that we'll end up with the best design for our situation, just a local optima. This explains why we still have blind spots, and a tendency for bad backs, it would require major structural reworking for us to be otherwise.

Yes, but I was trying to get T'ai to understand.

The same goes for things like computer keyboards, cities, etc.--the system can only build upon that which has evolved so far-- and what is there may not be the best "floorplan" for the future.

(And "local minimas" sounds so unfriendly...ecological niche has a nicer ring, I think.)

And "best" is relative. In terms of DNA it pretty much refers to survival value and reproductive fitness. Certainly humans have different ideas of what is "best". Randomness applies to the endless recombinations and alterations of genomes--Selection refers to the very few of these that survive--the "fittest" (and/or luckiest?) in any given locale stick around and get to stay in the "evolution" game.

[articulett]

Well that's clearly a lie to anyone that has read what I've written, and what others here have written.

It seems like the people who say things like 'well, evolution a stochastic process.... but it is not random!' are the ones denying reality, becase they have a strong emotional reaction to the word random, or a deep misunderstanding of its meaning.

Evolution is NOT random--mutations are random--recombinations involve randomness-- Selection is not.

The lotto balls are chosen at random. Occasionally there is a winner-- the pool of winners grows-- The lottos selects those who will be in the pool of "winners".

Get it? I thought not.

Let's try again--in a world of random events happening all the time to random people--some people end up extraordinarily unlucky...and some end up very lucky and we can expect some amazing coincidences. (And though we can expect such things in a world of activity with an input of energy no psychic seems able to predict any of them with any accuracy I might add.) We humans will notice these things because they stir up interest. But we will never see AAALLLLLLLLLLL the near misses where such things didn't happen from the experimental pool that reveals those that did. The fact that someone wins a lottery and someone dies in a freak accident is not at all unusual. But the human mind can invent meaning as to why such a thing happened to this or that person or even presume it happened for a "higher purpose". That does not make it so.

The genetic lottery is being played all the time (randomness). We only observe the winners (selection).

Why can't creationists EVER understand this simplistic concept?

[Foster Zygote]

Wow, such detailed arguments! Look, then just show me a non-trivial process with any random inputs whose result you can predict with 100% certainty. It's there waiting for you to do.

Way to move the goalpost. After Joobz provided you with ****'s 2nd law you now demand that evolutionary theory be able to provide you with an utterly ridiculous 100% predictability.

[Walter Wayne]

There are a few processes in the theory of evolution which could be described as random.And for a process to be random, it doesn't require that each subprocess is random. In fact the minimum requirement is one subprocess.
The chance of a specific locus having a Single Point Mutation can be determined to a set value but, much like flipping a coin, it might not happen. Mutations at individual loci is random, in that it cannot be predicted. However, the overall rate of mutation can be predicted extremely well, much like the overall number of "heads" and "tails" from a thousand coin flips can be extremely well predicted.
But coin flips are independent of each other. You can't predict 4 coin flips extremely well. But even if the all 4 came up heads, your new expect value of a thousand tosses, given this info on the first four, 502 (as opposed to 500 +/- 30ish) that well within our prediction.

When populations are small, the first few results don't get washed out in the long term. Subsequent generations bare that mark of the first generation "coin tosses".
Furthermore, there is often clear evidence that certain mutations are far more common then others. SPMs are not nearly as simple as "happens" or "doesn't happen". What's more, the mechanisms behind SPMs is fairly well understood. It might turn out to not be random at all. After all, in most cases it does not involve anything "quantum".The relative frequency not being equal doesn't make it non-random.

[Vorticity]

Here we go again with this mess of a topic. This has been discussed before:

Here:
http://forums.randi.org/showthread.php?t=50550

And here:
http://forums.randi.org/showthread.php?t=80924

In both those threads, I made my opinion clear, but I'll give it again here:

Evolution is indeed random (i.e. it is a nontrivial stochastic process), by the unambiguous, technical definitions of these terms*.

Keep in mind that saying this is in no way a criticism of evolutionary theory.

*I almost said "...but evolution is not 'random' by the layman's definition of the term', but then I realized that the layman's definition of 'random' is wholly incoherent.

[JoeTheJuggler]

I've got a drawer full of beads of different colors all jumbled up together, then I scoop out a big sample of them. Then, from my scoop, I select just the green beads and throw all the rest back in.

Is this process random?

[Walter Wayne]

I've got a drawer full of beads of different colors all jumbled up together, then I scoop out a big sample of them. Then, from my scoop, I select just the green beads and throw all the rest back in.

Is this process random?No.

But this is a very bad analogy to evolution. No heredity, no changing selection criteria, and absolute selection criteria as opposed to green beads be more likely to be selected.

An asides, in your experiment, if I asked you, "what colour are the beads left in your hand?" there are two possible answers; "Green" and "I have no beads left in my hands". So if I was being a pedant your original process yields a random colour, green or no colour.

Also, even if you can guarantee that some green beads make it into your scoop, you have chosen a trivial process. Regardless of T'ai's motive's, he did ask for a non-trivial process and the example you have is trivial (if you guarantee some greens make it to the scoop).

Walt

[Thabiguy]

I've got a drawer full of beads of different colors all jumbled up together, then I scoop out a big sample of them. Then, from my scoop, I select just the green beads and throw all the rest back in.

Is this process random?

It is non-random as far as the color of beads in your hand is concerned.
It is random as far as the number of beads in your hand is concerned.

Asking whether the process is random or non-random per se is like asking whether the surface of the Earth is solid or liquid.

[Ichneumonwasp]

Why is the fact that random processes underlie one part of evolution by means of natural selection such a big deal? The randomness occurs within an already existing structure and selection provides a direction (not pre-determined toward a particular goal). That randomness is a part of the process does not mean that the entire process is random. Such discussions are silly.

[CFLarsen]

Why is the fact that random processes underlie one part of evolution by means of natural selection such a big deal? The randomness occurs within an already existing structure and selection provides a direction (not pre-determined toward a particular goal). That randomness is a part of the process does not mean that the entire process is random. Such discussions are silly.

T'ai Chi is merely spouting one of his fellow Creationists' favorite strawmen:

Five Major Misconceptions about Evolution (http://www.talkorigins.org/faqs/faq-misconceptions.html)

"The theory of evolution says that life originated, and evolution proceeds, by random chance."

There is probably no other statement which is a better indication that the arguer doesn't understand evolution.

Ayup.

[Vorticity]

That randomness is a part of the process does not mean that the entire process is random.
In general, it does mean that the entire process is random... by the technical definition of 'random'.

The common creationist (or IDer, or whatever) argument seems to be something like:

1) Evolution is random.

2) Therefore evolutionary theory is invalid.

I think that people often attack the wrong piece of this argument. Instead of replying, "No, evolution is not random!", they should be replying "Sure, evolution is random. So what?"

[Ichneumonwasp]

In general, it does mean that the entire process is random... by the technical definition of 'random'.

The common creationist (or IDer, or whatever) argument seems to be something like:

1) Evolution is random.

2) Therefore evolutionary theory is invalid.

I think that people often attack the wrong piece of this argument. Instead of replying, "No, evolution is not random!", they should be replying "Sure, evolution is random. So what?"

Then the technical definition provides the wrong implication. By saying the entire process is random implies (at least to some) that all parts of it are random, which is not the case. The random changes occur on a matrix full of complex information. So the technical definition should be avoided since it can be improperly manipulated semantically, which seems to be creationists' favorite pasttime.

[John Hewitt]

Looks like another piece on randomness being involved in evolution:

http://scienceblogs.com/pharyngula/2006/03/chance_and_regularity_in_the_d.php

I think the article cited in the OP is very interesting but, so it seems to me, it is not about evolution, it is about development. There are many studies which point to the idea of development being guided by cellular waves, both waves formed within the cell and waves formed by the concerted action of arrays of cells. The emergent picture seems quite reminiscent of the Turing models for development.

I guess what I am saying is that this study, though interesting, doesn't point to a new situation or to one that is physically, chemically or biologically inexplicable.

[skeptigirl]

For once, T'ai Chi actually links to an article that really is cool and interesting.
I just hope he pays attention to the parts that talk about the developmental process being "unguided", instead of misinterpreting it as "special design", like some are prone to.It's a great piece. To the person less familiar with genetic science, it may appear to be something "designed", but the author correctly summarizes at the end how it all fits together and is not only compatible with evolution theory, it actually supports it.

Genetic science provides the blueprint for the structures and shows further evidence of how they developed a step at a time via the random accumulation of genetic mutations.

There's an additional piece in the picture I have mentioned before. Scientists experimentally (using evolution theory to predict, test, and confirm, aka the last step in confirming a theory) exchanged the rabbit gene which controls the development of the fetal eye with the fruit-fly gene that has the same function in a fruit fly larva. The result was a normally developed fruit fly eye in the fruit fly. This occurs despite the fact rabbits have mammal eyes and fruit flies have compound insect eyes.

The study of genes has revealed the segmented function where the gene that controls eye development in the fetus or larvae is separate from the gene that controls the eye structure and that in turn is separate from the gene that controls blood supply to the eye, nerves and so on. So just as the little pieces described in this article are simple, adding up to a complex whole, genetic research has revealed how that occurs in the evolutionary process.

[Walter Wayne]

Then the technical definition provides the wrong implication. By saying the entire process is random implies (at least to some) that all parts of it are random, which is not the case. The random changes occur on a matrix full of complex information. So the technical definition should be avoided since it can be improperly manipulated semantically, which seems to be creationists' favorite pasttime.
And if we use the layman's definition of random we are worse off, since there are so many different definitions. As I have pointed out, under many of those definitions evolution is still random.

The "yes it is random! no it isn't!" turns into a exercise in misinformation by both sides. In fact T'ai is arguing not only for the more accurate of the two conclusions by technical and many common definitions, but the argument he has presented in this thread is sound.

Walt

[skeptigirl]

I think the article cited in the OP is very interesting but, so it seems to me, it is not about evolution, it is about development. There are many studies which point to the idea of development being guided by cellular waves, both waves formed within the cell and waves formed by the concerted action of arrays of cells. The emergent picture seems quite reminiscent of the Turing models for development.

I guess what I am saying is that this study, though interesting, doesn't point to a new situation or to one that is physically, chemically or biologically inexplicable.
Your wave hypothesis is no further along in acceptance by the scientific community is it, John?

[skeptigirl]

Mutations are random, selection isn't. What is so difficult about that?

Oh, and variability itself is increased by selection.

[CFLarsen]

And if we use the layman's definition of random we are worse off, since there are so many different definitions. As I have pointed out, under many of those definitions evolution is still random.

The "yes it is random! no it isn't!" turns into a exercise in misinformation by both sides. In fact T'ai is arguing not only for the more accurate of the two conclusions by technical and many common definitions, but the argument he has presented in this thread is sound.

Walt

So, you disagree with the arguments in the TalkOrigins article?

[Ichneumonwasp]

And if we use the layman's definition of random we are worse off, since there are so many different definitions. As I have pointed out, under many of those definitions evolution is still random.

The "yes it is random! no it isn't!" turns into a exercise in misinformation by both sides. In fact T'ai is arguing not only for the more accurate of the two conclusions by technical and many common definitions, but the argument he has presented in this thread is sound.

Walt

But the whole idea is wrong-headed. Evolution is not one thing. There is mutation, which is random. Mutation occurs on complex information, which is not random. And natural selection is not random. Plus there are many mechanisms underlying evolution that are not covered by these terms. So, technically, if we have a process that includes one random item but is largely not random (as a whole), why call the whole thing random when it gives the wrong impression? When discussing the matter we should be more precise so as not to engage in needless semantic shuffles that only obscures the real issues.

Call mutation random, which is what it is. Call the underlying genetic information a non-random collection of useful information. Call natural selection a largely non-random event the provides direction for the process but no set goal. Then we needn't be subjected to idiotic 'metaphors' such as how could a tornado create a 747 (which is wrong on more than one level).

In a sense, saying that evolution is random is like me saying that anemia is caused by iron deficiency. Well, yeah, there is one form of anemia caused by iron deficiency, but there are many other forms of it where iron levels may be completely normal and there is a gene defect or B12 deficiency, etc. Like anemia, 'evolution' is a blanket term.

ETA

And we cannot call mutation completely random. It follows some rules, afterall. It is not predictable, but it cannot be anything at all. Either there is a point mutation, which must assume a particular form, a frameshift, a deletion; gene duplication, etc. There are many ways of interpreting the word 'random', so we should be more precise in our discussions of such terms.

[articulett]

In general, it does mean that the entire process is random... by the technical definition of 'random'.

The common creationist (or IDer, or whatever) argument seems to be something like:

1) Evolution is random.

2) Therefore evolutionary theory is invalid.

I think that people often attack the wrong piece of this argument. Instead of replying, "No, evolution is not random!", they should be replying "Sure, evolution is random. So what?"

Because it's not random. Mutation and recombination have random components--evolution is the direction selected from amongst the successful mutations and recombinations via the environment.

If you are trapped in a cave and decide to dig your way out...you might choose where to dig at random...but chances are, you will continue to dig in that direction until it starts to feel like a lost cause or another way seems to be better, etc. The pattern you make while digging will evolve (change through time)--your first choice of where to dig may have been random...but each successive etching away of wall will be based upon what you've done already--you are selecting. The environment selects. If a million people were digging their way out of caves and only those who made it out alive survived to reproduce--then whatever genes that may have facilitated their survival will be "selected for" in the genome of future offspring.

Mutations are "random" (though not completely)
Recombinations are "random" (though not completely)
Selection is not. Evolution is not. The latter is more important than the former in the evolutionary process. Each evolutionary "step" had to only happen one time in the plethora of ongoing ever increasing DNA-copying going on. Each genome is a record of those steps.

[Foster Zygote]

In general, it does mean that the entire process is random... by the technical definition of 'random'.

The common creationist (or IDer, or whatever) argument seems to be something like:

1) Evolution is random.

2) Therefore evolutionary theory is invalid.

I think that people often attack the wrong piece of this argument. Instead of replying, "No, evolution is not random!", they should be replying "Sure, evolution is random. So what?"

I agree with you... mostly. The problem with saying that would be that most creationists will then claim victory in the belief that evolution is "random" in the sense that throwing a box of parts out of a plane and expecting them to assemble into a working pocket watch on impact is "random". I think that a better thing to say would be "Yes, evolution is partly random. But your understanding of random as it applies to evolution is incorrect". It's similar to something I already say often: "Yes, evolution is a theory, but "theory" doesn't mean what you think it means."

[articulett]

It's a great piece. To the person less familiar with genetic science, it may appear to be something "designed", but the author correctly summarizes at the end how it all fits together and is not only compatible with evolution theory, it actually supports it.

Genetic science provides the blueprint for the structures and shows further evidence of how they developed a step at a time via the random accumulation of genetic mutations.

There's an additional piece in the picture I have mentioned before. Scientists experimentally (using evolution theory to predict, test, and confirm, aka the last step in confirming a theory) exchanged the rabbit gene which controls the development of the fetal eye with the fruit-fly gene that has the same function in a fruit fly larva. The result was a normally developed fruit fly eye in the fruit fly. This occurs despite the fact rabbits have mammal eyes and fruit flies have compound insect eyes.

The study of genes has revealed the segmented function where the gene that controls eye development in the fetus or larvae is separate from the gene that controls the eye structure and that in turn is separate from the gene that controls blood supply to the eye, nerves and so on. So just as the little pieces described in this article are simple, adding up to a complex whole, genetic research has revealed how that occurs in the evolutionary process.

I think this is so cool too...it's like the gene is a pattern for "eyes"--and it can be modified according to which type of eye works best in an organism via "instructions" from other genes in the genome. I remember how cool it was to discover the hox genes. It just makes such sense. Sure it's humbling. But storybooks and great literature and everything in between are written with the same 26 letters (in English) and all of life is built using the same 4 basic chemicals...even fruit fly life. Useful templates are conserved and modified as needed.

[Walter Wayne]

But the whole idea is wrong-headed. Evolution is not one thing. There is mutation, which is random.
Yes.
Mutation occurs on complex information, which is not random.
I disagree with this. In fact you are stating your conclusion here. If evolution is not random process then the [genetic] information is not random, if evolution is random, then the information will be random.
And natural selection is not random.
A non-random process working on random information usually produces random information. As such the non-randomness of selection alone establishes nothing.
Plus there are many mechanisms underlying evolution that are not covered by these terms. So, technically, if we have a process that includes one random item but is largely not random (as a whole), why call the whole thing random when it gives the wrong impression? When discussing the matter we should be more precise so as not to engage in needless semantic shuffles that only obscures the real issues.
The inclusion of more mechanisms doesn't necessarily lead to a less random system. In fact many differing factors acting on a system simultaneously could lead to a rather chaotic system. Throw a random input into a chaotic system and the output is random in both the technical and colloquial senses of the word.
Call mutation random, which is what it is. Call the underlying genetic information a non-random collection of useful information. Call natural selection a largely non-random event the provides direction for the process but no set goal.
How does putting random mutations into a selection process generate random data? Especially in those periods when populations are small, so you can't guarantee a "law of large numbers" type situation.

The complexity of the mechanism only exagerate the randomness. Suppose a small population, where the randomness is most noticeable, survives and the population grows and spreads. The variation not only affects that species, but it shapes the environment for the animals around it when the population expands. The animals that prey on it, are preyed on by it, or live with it (parasitic or symbiotic) all have their environments shaped by it.

Not only does a species have to live with its genetic history, so to does every species whose "evolutionary environment" is affected by it. A period of a freak random result doesn't get forgotten. It stays with us. And how long has evolution gone on? How many chances for small populations to go the less likely route.
Then we needn't be subjected to idiotic 'metaphors' such as how could a tornado create a 747 (which is wrong on more than one level).
The solution is to combat there understanding of random systems (and once they learn that, we'll be subjected to a new idiotic metaphor from the close-minded.)
In a sense, saying that evolution is random is like me saying that anemia is caused by iron deficiency. Well, yeah, there is one form of anemia caused by iron deficiency, but there are many other forms of it where iron levels may be completely normal and there is a gene defect or B12 deficiency, etc. Like anemia, 'evolution' is a blanket term.
At any point the evolution pressures on a species is affected by the mutations that occured and survived in it and other species around it.

Your anemia has no affect on my anemia.

ETA

And we cannot call mutation completely random. It follows some rules, afterall. It is not predictable, but it cannot be anything at all. Either there is a point mutation, which must assume a particular form, a frameshift, a deletion; gene duplication, etc. There are many ways of interpreting the word 'random', so we should be more precise in our discussions of such terms.
Responding to the part I bolded, I've never rolled a 7 on a six-sided die, or flipped Lincoln's head on a canadian penny. Able to be "anything at all" is not part of the technical or common definitions of random.

Walt

[cyborg]

The assertion that any non-determinism in a system makes the entire system non-deterministic is plain wrong. There are mathematically proven systems that have non-deterministic components that are nonetheless overall deterministic.

So anyone using that argument as an establishment that evolution is inherently random is inherently wrong.

That is all. Carry on.

[Thabiguy]

Walter,

can you name a real-life example of a process that is not technically random?

[Walter Wayne]

The assertion that any non-determinism in a system makes the entire system non-deterministic is plain wrong. There are mathematically proven systems that have non-deterministic components that are nonetheless overall deterministic.

So anyone using that argument as an establishment that evolution is inherently random is inherently wrong.

That is all. Carry on.As is the other way round. And if you notice I mention other traits other than just the non-determinism of mutation in my arguments, just to head off the millions of coin flips arguments including hereditary reproduction, periods of small breeding populations and the affect of each species on the others around it.

Would you care to give me a mathematically proven system that have non-deterministic components that are nonetheless overall deterministic?

Then can you tell me how it applies as a good analogy for evolution?

Walt

[Walter Wayne]

Walter,

can you name a real-life example of a process that is not technically random?At some level, depending on how close to 0 you like your 0's to be, everything is technically random. Though most technical people would only qualify those examples with "but determistic for all intents and purposes".

As the old "scientific warning labels" e-mail that went around said
Advisory: There is an Extremely Small but Nonzero Chance That, Through a Process Known as 'Tunneling,' This Product May Spontaneously Disappear from Its Present Location and Reappear at Any Random Place in the Universe, Including Your neighbor's Domicile. The Manufacturer Will Not Be Responsible for Any Damages or Inconvenience That May Result.
However, even a pedant would not claim that the process whereby my desk doesn't appear in your home tomorrow, without passing through the points in between is random except in the strictest most useless definition of the term.

However, I have argued that not only is evolution random in the technical sense, but that it is random in most non-technical senses as well. I've also think that it is very strongly random. By that I mean that given the universe in the precise position it was once life had established itself on this planet, I don't think humans, intelligent life, or even mammals as we known them are a necessary result if we replay it, even throwing in the odd meteor and such at the same times as before. I am not saying it is "weakly random" in the sense that a strong electrical signal has a small noise component that easily comes out in the wash.

The long term randomness is due to the short term randomness, not some indeterminancy at beginning.

Walt

[cyborg]

Would you care to give me a mathematically proven system that have non-deterministic components that are nonetheless overall deterministic?

The Network Algebra for dataflow/stream processing languages in computing makes use has non-deterministic data flow with deterministic data processing units that produces a network that is deterministic overall.

Then can you tell me how it applies as a good analogy for evolution?

The operation of evolution is probably best described as the process of exploring a genetic search space with non-deterministic branching - you could replace the non-deterministic search with a deterministic one and the acceptance of the candidate organisms would not be affected. Of course the situation is somewhat complicated by the acceptance function being dynamically updated by the solutions that are accepted.

[Thabiguy]

I've also think that it is very strongly random. By that I mean that given the universe in the precise position it was once life had established itself on this planet, I don't think humans, intelligent life, or even mammals as we known them are a necessary result if we replay it, even throwing in the odd meteor and such at the same times as before. I am not saying it is "weakly random" in the sense that a strong electrical signal has a small noise component that easily comes out in the wash.

The long term randomness is due to the short term randomness, not some indeterminancy at beginning.

Are you saying that you believe that evolution is chaotic, i.e. arbitrarily small changes in initial conditions will eventually lead to significantly different future outcomes?

[Taffer]

And for a process to be random, it doesn't require that each subprocess is random. In fact the minimum requirement is one subprocess.

Then your use of "random" differs from mine.

But coin flips are independent of each other. You can't predict 4 coin flips extremely well. But even if the all 4 came up heads, your new expect value of a thousand tosses, given this info on the first four, 502 (as opposed to 500 +/- 30ish) that well within our prediction.

Do you know how the molecular clock works?

When populations are small, the first few results don't get washed out in the long term. Subsequent generations bare that mark of the first generation "coin tosses".

And this is called genetic drift. And it can be modelled.

The relative frequency not being equal doesn't make it non-random.

That's not what I said. I said we don't know for certain that mutation truely is random, and cited differing mutation ratios as an example which lead to this thought.

[CFLarsen]

Walter Wayne,

So, you disagree with the arguments in the TalkOrigins article?

[John Hewitt]

Your wave hypothesis is no further along in acceptance by the scientific community is it, John?
No, not so far as I know, but I don't do "the bang head on brick wall" thing.

My present understanding is that the cell motility field never mentions the word "wave" and holds that there is no evidence to suport their existence and/or relevance. On the other hand, the development field writes reviews about them and holds meetings to discuss their importance - for my own background, I went to one a couple of months back.

[Ichneumonwasp]

Yes.

I'm not sure what definition of 'random' you are using, so let's be clear. The 'yes' to mutations are random is an epistemic 'yes', correct? It does not mean non-deterministic since we could potentially predict mutations if we had more knowledge. For the most part mutations arise through copy errors.

I disagree with this. In fact you are stating your conclusion here. If evolution is not random process then the [genetic] information is not random, if evolution is random, then the information will be random.

I'm not offering a formal argument, so I don't see how I could be stating my conclusion. We often define information to mean 'not-random'. What does 'random information' mean? It doesn't matter to the process how the information came into place, the issue is that it is relatively stable so that further changes can build upon that base. The argument is against the typical creationist straw man use of 'random' to mean completely chaotic -- with the implication that we must rebuild the entire structure from the beginning. The way that many creationists twist the word 'random', they imply that no information could ever arise from it. That is what we argue against. That is why the way you are using the word here might create more problems in the short term.

A non-random process working on random information usually produces random information. As such the non-randomness of selection alone establishes nothing.

Using 'random' in that way might hurt arguments in the short term. As used in most arguments, 'random' means 'absence of information', so 'random information' is an oxymoron to creationists. I am not arguing against formal definitions of random, but the way the word is used in this debate. Words are often twisted, so giving folks more ammunition is a bad idea in my book.

The inclusion of more mechanisms doesn't necessarily lead to a less random system. In fact many differing factors acting on a system simultaneously could lead to a rather chaotic system. Throw a random input into a chaotic system and the output is random in both the technical and colloquial senses of the word.

All fine as far as it goes. But evolution is not a purely chaotic system. There is structure on which further changes scaffold.

Not only does a species have to live with its genetic history, so to does every species whose "evolutionary environment" is affected by it. A period of a freak random result doesn't get forgotten. It stays with us. And how long has evolution gone on? How many chances for small populations to go the less likely route.

Yes, that is the whole point and why I harp on it. Since creationists often use the word 'random' to mean 'chaotic', this issue points to the preservation of 'random changes'. That is the whole point of discussing selection and persistant genetic information. Getting a creationist to agree to your definition of 'random' is fine with me if you can get them to do it. All I have ever seen them do is try to manipulate it.

The solution is to combat there understanding of random systems (and once they learn that, we'll be subjected to a new idiotic metaphor from the close-minded.)

O.K., I think we agree. I don't see any hope for getting creationists to use a formal definition of random and stick to it, though. I have been arguing with Kleinman recently and all I can see of his technique is that he moves definitions, intentionally misinterprets information and posters' stances, and refuses to answer any points made against him. While I find it noble that you want to formalize the definition to make the debate less slippery, I'm becoming more jaundiced concerning the potential benefit.

Responding to the part I bolded, I've never rolled a 7 on a six-sided die, or flipped Lincoln's head on a canadian penny. Able to be "anything at all" is not part of the technical or common definitions of random.

Walt

That's fine. I was only trying to ensure that you weren't using a creationist type definition of the word 'random' that at times seems to mean 'purely chaotic' and at other times 'anything can happen' and at other times 'non-deterministic', etc. I'm afraid that I have seen the 'anything at all' definition thrown about by creationists in the past.

[aggle-rithm]

As is the other way round. And if you notice I mention other traits other than just the non-determinism of mutation in my arguments, just to head off the millions of coin flips arguments including hereditary reproduction, periods of small breeding populations and the affect of each species on the others around it.

Would you care to give me a mathematically proven system that have non-deterministic components that are nonetheless overall deterministic?

Then can you tell me how it applies as a good analogy for evolution?

Walt

Evolution is not deterministic, but there are limits to what can be expected by the process. For instance, non-viable structures such as a cell without a membrane, or a large organism without some sort of supporting structure can't be expected to result from evolution.

Saying a process has to be either completely random or completely deterministic is a false dilemma. For instance, computer programs that select from a limited set of objects based on random numbers are fairly common. An example would be a simple program that displays fortune cookie messages. There would be no way to predict which fortune would appear next (here we are assuming a mythical program with a true random-number generator), but that doesn't mean the process is completely random; if it were, then the vast majority of the time the "fortune" would be jibberish. In this, case, a random number is used to select from an existing pool of choices. Evolution is more complex than this, but this analogy is a lot closer to reality than that of the "coin flip".

[Belz...]

Another pointless comment. Why not try being a real skeptic and talk about the article?

Do you ever actually discuss in the threads you start ?

I thought the randomness is quite interesting, since many drone on about how evolution is not random.

Depends what you mean by "random".

It is a conception based off of what Dawkins, Miller, and many others, who say that evolution is the selection of mutations and other things.

Well, it is, in part.

[Belz...]

Would you care to give me a mathematically proven system that have non-deterministic components that are nonetheless overall deterministic?

Random quantum fluctuations aside because we can't tell if they have an actual impact or not, there is no such thing as randomness; only things we can't predict because of insufficient data and processing power.

[Tief]

It is a conception based off of what Dawkins, Miller, and many others, who say that evolution is the selection of mutations and other things.

So you are saying this formula of yours came out of mathematical formulas by these people?

Ah, since it is a conception based upon what Dawkins said, I must accept it as verifiably a mathematical proof.

Sorry, I only made it through Dif EQ, and that was 30 years ago, you're going to have to spell out in small terms exactly why your formula is more than just wishful thinking. I expected this when I followed your link, but was greatly disappointed.

[Walter Wayne]

I haven't had time to post much recently, so I apologize for the late reply.
The Network Algebra for dataflow/stream processing languages in computing makes use has non-deterministic data flow with deterministic data processing units that produces a network that is deterministic overall.
I now little of that subject and it has been about 10 years since I did any dataflow/network analysis, and that was school work. I can't really comment on that with the knowledge I have now.
The operation of evolution is probably best described as the process of exploring a genetic search space with non-deterministic branching - you could replace the non-deterministic search with a deterministic one and the acceptance of the candidate organisms would not be affected. Of course the situation is somewhat complicated by the acceptance function being dynamically updated by the solutions that are accepted.
I think the relevant differences between this and evolution are many, and are the reason why such algorithms converge.

For one, the "right" solution doesn't change. In biological evolution you have a changing environment, just look how the mammal "solution" changed after the age of dinosaurs. That environment also changes depending on other species around you. In your genetic search space the minima are static. In fact the reason you have a solution is because it is static. Some of the "good solutions" in biological evolution today likely sucked for ice-age, and the reverse is also true. If you have a non-determistic path, and the very nature of your solution space changes with time you won't converge to a final solution.

Another point is competing resources. Imagine that each iteration of your algorithm you have the time to compute 10 points in the search space. If you find a local minima that is "good enough" but not perfect you can through it out, or if you find a solution, you don't recompute in future generations. If your search space has 11 solutions, you can still find them all because once you find one, you don't have to spend resources on it anymore.

That is not true of biology where a species that is fit, continues to draw resources from the environment. Mother nature doesn't say, "well species X is adequate, but they have all these little flaws so I'll just scrap them and use the resources they took up to build a new species in that niche."

So back to alogorithm where in each iteration you can compute 10 points, but the search space has 11 solutions. If with each of iteration you are forced to dedicate resources (computer cycles) to solutions you have already found your algorithm will find at most 10 solutions in that space. And because the branching is non-determistic, neither is the solution you miss.

By a very simple adjustment, the algorithm becomes non-determistic.

Walt

[Walter Wayne]

Are you saying that you believe that evolution is chaotic, i.e. arbitrarily small changes in initial conditions will eventually lead to significantly different future outcomes?
Do I think evolution is chaotic? No, I think it is random (technically, and by most non-technical definitions). I think give the same starting conditions can lead to different outcomes.

Do I think selection is chaotic?
Been awhile since a looked at chaos theory so I will give a yes, but provisional on my remembered knowledge being correct.

I would put in a few other changes in the statement. First I wouldn't say arbitrarily small changes. Second I wouldn't say initial conditions, I would say inputs. An important distinction since mutations (the random element) continue to occur, where as the initial condition would be t=abiogenesis so to speak.

Let's state my opinion as changes of inputs on the magnitude causable by mutation can lead to significantly different outcomes from selection. Evolution being a process composed of mutation (the random input) and various selection processes is thus random.

Any example I give will be somewhat forced given the complexity of an ecosystem but I will try to provide one that illustrates my point. Suppose a species comes under pressure due to shrinking habitat (a natural disastor perhaps), and during that time there is only enough room in the ecosystem for a small population. To introduce a random element, around that time a disease mutates such that it is able to jump to that species and devastates the population. That random bit is that the disease mutation may occur randomly a generation or two prior to the species coming under pressure from the disastor I mentioned previously, or during that period when the population has already shrunk.

In situation 1, where the disease devastates the population prior to their habitat shrinking you get massive reduction in population from the disease, but the smaller population isn't affect as much by reduction in habitat because they are few in number already. You end up with a small population, most of which have already been exposed to and survived the disease.

In situation 2, where the disease jumps to the species after the habitat reduction you get a different result. After the habitat reduction the population shrinks to a small number as above, but they have not yet been exposed to the disease. When it hits, the already vulnerable population shrinks even further.

Now in situation 2 the population may die out, or when habitat opens up again it may not have the numbers to compete for niches with other more healthy competitors. If it doesn't succed that cascades through the ecosystem as that species is an evolutionary pressure on the species around it.

As I mentioned before, in large populations the random variations may even out over time. With 6 billion humans, many of whom are somewhat "indescrimate" in selection of a mate there are few isolated small breeding populations. But in species underpressure, or especially in those points in time where entire ecosystems are under huge pressure those random variations can have a lasting affect.

The example I chose had a rather immediate impact. But with species being so interdependent on each other, I doubt the differences have to be as distinct as whether one species lives or dies out to have a notable affect on the long term future.

Walt

Edited to add: Wow, I didn't expect that reply to be so long. I'll try to get back to the other question later.

[cohen avshalom]

could the universe began with complete zero-no mass/energy?
second question of me:could area change state,like any other material? from solid(state)->liquid(state) ->gas(state).i mean the area himself the shield-even with full vacuum.

i was writting 3 pages that call icarus5.
you can see them in "icarus5-com "
i hope you will note me back-write in easy way to understand-for everyone.i wait for your opinion.

**********************
thank you from:
cohen avshalom charly
isreal/haifa

**********************

[aggle-rithm]

could the universe began with complete zero-no mass/energy?
second question of me:could area change state,like any other material? from solid(state)->liquid(state) ->gas(state).i mean the area himself the shield-even with full vacuum.

i was writting 3 pages that call icarus5.
you can see them in "icarus5-com "
i hope you will note me back-write in easy way to understand-for everyone.i wait for your opinion.

**********************
thank you from:
cohen avshalom charly
isreal/haifa

**********************

Why do people always confuse cosmology with evolution? They're nothing alike.

*Sigh*

[Soapy Sam]

Hey- he's confused about the reality of Israel.

[Tanstaafl]

Turns out he's just spamming.

Identical posts in many different places, and twice within the JREF forums.

[six7s]

I know for sure that forms like the elephant won't grow wings and develop flight.

Has PZ been reading this old thread?

scienceblogs.com/pharyngula Elephants’ wings (http://scienceblogs.com/pharyngula/2009/05/elephants_wings.php)
Category: Godlessness
Posted on: May 10, 2009 2:31 PM, by PZ Myers

Once upon a time, four blind men were walking in the forest, and they bumped into an elephant.

Moe was in front, and found himself holding the trunk. "It has a tentacle," he said. "I think we have found a giant squid!"

Larry bumped into the side of the elephant. "It's a wall," he said, "A big, bristly wall."

Curly, at the back, touched the tail. "It's nothing to worry about, nothing but a piece of rope dangling in the trail."

Eagletosh saw the interruption as an opportunity to sit in the shade beneath a tree and relax. "It is my considered opinion," he said, "that whatever it is has feathers. Beautiful iridescent feathers of many hues."

<snip/>

Eagletosh yawns and stretches in the shade of a tree. "It has wings, large wings, that it may ascend into the heavens and inspire humanity. There could be no purpose to such an animal without an ability to loft a metaphor and give us something to which we might aspire."

The other three ignore the idling philosopher, because exciting things are happening with their elephant!

<snip/>