What evidence is there for evolution being non-random?

[mijopaalmc]

So you will not accept anything I present as evidence for my argument, articulett?

Could you also explain how a stochastic process is not random?

 

[Meadmaker]

I guess as a side question, what is everyone's favorite term to use when describing evolution/natural selection/speciation?

If I had to pick one it would be "complex".

Oddly enough, although I am one of the primary "randomites", I, too would usually avoid "random". What interested me, though, was that Dawkins didn't merely avoid "random"; he absolutely insisted that it wasn't correct. He called evolution "the exact opposite of chance", and devoted quite a few words to the idea. It seemed to be a rather visceral, emotional, dislike, not unlike what we have seen in this thread. While it isn't always the best word, it is occasionally appropriate, depending on the audience and the specific topic, and I found it interesting that the use of that word elicited such a strong negative response.

I still do, and I still think that it has nothing to do with science.

 

[steenkh]

Actually this doesn't appear to happen when you look at the population as a whole, rather than just a sample that you have collected in the field.

There is some confusion here because we often use the word "mutation" but fitness is rarely linked to a single mutation but to the entire genome. You can have mutations that lower your survivability in one respect, but the other genes can still give you a sufficient advantage so that you are still 'fit'.

And, as you also note, the fitness will have to be judged over a longer time period to compensate for random flukes, and over a larger population, because single unfit genomes can sometimes survive, but not in a whole population.

I have not read the study you referred to, but I do not think I would have any quibble with it, as per the comments I made above. Your conclusion that "ordered behavior can arise form a random/stochastic process" is enough for me to conclude that evolution is non-random.

 

[articulett]

So you will not accept anything I present as evidence for my argument, articulett?

Could you also explain how a stochastic process is not random?

Hey, I'd accept lots of stuff, and I keep asking for it--and I'd accept it way more quickly than you accepted my evidence of similar nature. Do you have any peer reviewed papers that say "Evolution IS random"? How about papers defining random as you do in the text? How about any saying "natural selection IS random"? How about if one said, "natural selection is not determined" or "natural selection is best model as a random process". Or even something that says "any process containing any randomness can be called a stochastic process"? Maybe you can find one that says "random is a synonym for stochastic". Don't interpret the papers for me. I can understand them quite well myself, and they are not saying what you are trying to pretend they are saying.

I already accept your argument if all you are saying is that "natural selection has probable elements"...which means your OP question could be rephrased as "what is the evidence for evolution not having probable elements..." But I can't imagine you'd actually ask such a meaningless question. And you later said you wanted to know why Dawkins would call selection "non-random".

And I've said this all before about stochastic processes... Stochastic processes contain random variables, but they are not considered random in themselves... random components do not a random process make (unless your goal is to confuse.) A mathematical function has random variables too. But a function is not random in itself.

I provided exactly what you asked for in your OP--multiple credible sources explaining why natural selection is considered "non-random"-- many people came to this thread and told you the exact same information. But none of their information matters...because the only thing you meant by random is "of or relating to a probability chart". So the evolution of everything IS random... and all processes except mathematic formulas are also random. In fact, I'd be hard pressed to find anything that wasn't random per your loose definition. So maybe you should have been more honest from the get go: "Gee, I just want to know if there is anything about evolution that can't be described by a probability chart so I can say that scientists think this all came about "randomly". . .

No scientist of credibility will say that evolution IS random. Many will say that natural selection is the opposite of random. No scientist is using random as loosely as you are despite your pretenses to the contrary. And there is no evidence that will allow you to concede this because you are a creationists...and you need to believe that scientists think this all came about randomly. Heck, you cannot admit that random is a poor word choice even...your goal is to play whatever semantic games you need to play so that you can sum up evolution as random.

Evolution isn't an argument. It's a fact. And the best way to describe it will evolve. And it won't be evolving "randomly"--rather it will be biased towards undoing creationist attempts at muddling understanding. You are driving the evolution of such a description--but from the wrong side of the equation as far as I'm concerned.

 

[articulett]

If I had to pick one it would be "complex".

Oddly enough, although I am one of the primary "randomites", I, too would usually avoid "random". What interested me, though, was that Dawkins didn't merely avoid "random"; he absolutely insisted that it wasn't correct. He called evolution "the exact opposite of chance", and devoted quite a few words to the idea. It seemed to be a rather visceral, emotional, dislike, not unlike what we have seen in this thread. While it isn't always the best word, it is occasionally appropriate, depending on the audience and the specific topic, and I found it interesting that the use of that word elicited such a strong negative response.

I still do, and I still think that it has nothing to do with science.

Gee...and you still can't understand the reaction after dealing with rttjc....Imagine thousands of him organized to try and make sure people don't understand natural selection... imagine that you understand this stuff better than anyone and you just want to share the information so that others can understand it too...

Also, in this link http://www.msnbc.msn.com/id/19733274 I think we'd both agree the mutation was random... but the way in which it rapidly spread through the species was not random was it? Or would you actually say it was random? Wouldn't a person aiming for clarity be more correct in saying that the selection was "nonrandom". That's not politics. That's just humans doing the best they can to convey FACTS when a huge group of people have a vested interest in people not understanding those facts.

The thing is, you think you have some sort of intuitive feel for what creationists do and say and who is and isn't one....and you think you understand how natural selection works and that the real problem is that people don't understand random...

But you have overestimated all the above... and random IS an ambiguous term and a word used again and again to spread misinformation and obfuscate understanding of a really important and simple concept. Sure, evolution is complex... so is Google's algorithm. But the results aren't "random". Nor are the results of natural selection. And you wouldn't convey anything about how Google results are selected if you called it a "random process". The same goes for natural selection, and by corollary, evolution--because having random components does not make an entire process random. (Poker and roulette are not equally random).

 

[sphenisc]

I guess as a side question, what is everyone's favorite term to use when describing evolution/natural selection/speciation?

If I had to pick one it would be "complex".

Oh, "complex" ay?! Everyone knows mathematicians use 'complex' to mean having an imaginary component. You're trying obfuscate and confuse the meanings to imply that evolution's imaginary aren't you!!! Creationist, ay?!

 

[articulett]

Evolution IS complex--but the basics are pretty simple.

What term would you use to convey how the order arises from the pool of randomness and multiplies exponentially?

I think I'd say:

The best replicators built by the DNA amongst all the permutations pass their info. on preferentially.

 

[mijopaalmc]

Each and every article that models evolution by natural selection as a stochastic process implies that it is random. How you have managed to miss that point eludes me, but you have and you have also failed to provide non-cherry-picked quotations from the literature that I have provided that refute my analysis. Furthermore, Schneibster, jimbob, and I have repeatedly shown you that every source that all the sources you have cited including Dawkins contain a inherent contradiction in so far as the insist that evolution is "non-random" and then talk about how adaptations increase the "probability of survival and reproduction.

So, I challenge anyone to come up with a reason why referring to evolution as "random", while unwise because of the word's wide lexical range, is by definition incorrect without resorting to the straw that "random" in the sense of "[o]f or relating to a type of circumstance or event that is described by a probability distribution" (and only that sense) can describe everything.

 

[jimbob]

Nice explanation, but he's had a million of them. He cannot compute anything that allows for the comprehension that some part of evolution might be "non-random".
People are upset with me because I've called him (Mijo) an "intelligent design proponent".

mijopaalmc, what do you think scientists mean when they say that evolution is non-random? What do you think they are using non-random to mean in the context of evolutionary theory?

Something akin to evolution being a directed process that results in order, but for the millionth time the multitudes of limit theorems in probability and statistics imply that this is a result of multiple sampling events, not some underlying determinism to the process itself.

I hope this demonstrates that I am not trying to undermine evolution by saying that it is "random", but that the perceived order of evolution by natural selection is a result of the repetitious nature of the process, not inherent determinism.

Doesn't seem like an ID proponent to me.

An intelligen't designer needn't be a God.

Spot the difference.

Articulett, I recall you posting somewjhere on this thread about confrontational styles and needing to "win" arguments.

are you sure that isn't what you are doing?

What would mijo need to say to convince you that he was not advocating intelligent design? He is quite happy to use the term "stochastic", to avoid the implications of "haphazard" and "unbiased chance".

What mijo is saying is extremely clear to a numerate person.

Describing something as random doens't mean that every event has equal probability, even in everyday parlance. Otherwise bookmakers would offer equal odds on every outcome, because nobody would be able to understand anything else.

 

[Meadmaker]

Oh, "complex" ay?! Everyone knows mathematicians use 'complex' to mean having an imaginary component. You're trying obfuscate and confuse the meanings to imply that evolution's imaginary aren't you!!! Creationist, ay?!

 

[jimbob]

There is some confusion here because we often use the word "mutation" but fitness is rarely linked to a single mutation but to the entire genome. You can have mutations that lower your survivability in one respect, but the other genes can still give you a sufficient advantage so that you are still 'fit'.

And, as you also note, the fitness will have to be judged over a longer time period to compensate for random flukes, and over a larger population, because single unfit genomes can sometimes survive, but not in a whole population.

I have not read the study you referred to, but I do not think I would have any quibble with it, as per the comments I made above. Your conclusion that "ordered behavior can arise form a random/stochastic process" is enough for me to conclude that evolution is non-random.

My point is that to understand whether it is appropriate to simplify to a "nonrandom" approximation, you need to have an analysis of the probabilities.

This means the level of the mutation in the population, the reproductive advantage and the reproductive probabilities for the mutation's alleles.

Otherwise, you can say it is probabilistic, but not how.

The reproductive advantage of the mutation/trait can (in principle) be analysed at a population level, with statistical techniques.

In my previous hypothetical treatment of a trait conferring a 10% reproductive advantage, I assumed that the population was stable. If the population oscillates (e.g. lemmings) then a beneficial mutation would have to be lucky, or very advatageous (several thousand percent) to propagate if it occured during a crash period, but unlucky not to propagate if it occured during the growth period.

Those statements hold true whenever populations are growing or shrinking.

But steenkh, as you said before, in general natural selection is highly efficient at weeding out the less fit. I would say this follows from Malthus: populations grow until limited. When limited, given the initial size of broods, most individuals in a population will not reproduce.

If a brood size is ten per parent, and on average one offspring per parent reproduces, then each individual has a 90.4% chance failing to reproduce. If the mutation confers a massive 50% improvement in reproductive success, then this falls to a "mere" 86%. Of course if the individual with the mutation did breed (again asexually for simplicity) then there is a 23% chance of none of its offspring reproducing.

 

[jimbob]

So then it's random as to whether two carriers of sickle trait have a child with sickle cell anemia? And it's random as to whether you get an A on a test? And it's random as to whether seat belts save lives? -- and what passes through a sieve is random. If you have a royal flush, your chances of winning are random. That's just a piss-poor non-descriptive vague use of random. If you want to convey how order comes from randomness you better quit describing random as "any model that requires probability"... or provide at least one peer reviewed paper that describes random in that way.

So then it's random as to whether two carriers of sickle trait have a child with sickle cell anemia?

Essentially, yes. As you know the odds are 50% per child (assuming no screening), so the odds depend on the number of children.

If the family has 1 child the odds are 50% (1/2)
If the family has 2 children the odds are 25% (1/2)^2
If the family has 3 children the odds are 12.5% (1/2)^3

That none of the children will have sickle cell anaemia.
and so on.

How would you describe the chances of having a child with sickel cell anaemia


If only one parent was a carrier, then the odds are vanishingly that any child would have sickle cell anaemia. (The advantages of sexual reproduction and avoidance of inbreeding).

 

[articulett]

Doesn't seem like an ID proponent to me.

Spot the difference.

Articulett, I recall you posting somewjhere on this thread about confrontational styles and needing to "win" arguments.

are you sure that isn't what you are doing?

What would mijo need to say to convince you that he was not advocating intelligent design? He is quite happy to use the term "stochastic", to avoid the implications of "haphazard" and "unbiased chance".

What mijo is saying is extremely clear to a numerate person.

Describing something as random doens't mean that every event has equal probability, even in everyday parlance. Otherwise bookmakers would offer equal odds on every outcome, because nobody would be able to understand anything else.

Let's see... I gave tons and tons of evidence I would accept...heck, just getting him to admit that defining evolution in terms or randomness is misleading would be nice...anything at all which distinguishes his argument from Behe's would be nice. It's just that you guys don't really know how "intelligent design" proponents operate. They never admit to being a creationist or even an intelligent design proponent...instead they answer obliquely. They ask insincere questions that they really don't want answered. They show very little interest in the links and information they specifically asked for and a complete lack of curiosity as to what is going on in the world of science. Mijo has made this an issue about how you can sum up evolution as random. That is exactly what creationists do. This looked like a thread of a person wanting to know the ways evolution was not random--why Dawkins would say that natural selection is not random... asking for peer reviewed papers that say evolution is not random. Well there aren't going to be any peer reviewed papers that evolution is random or non-random because random is an ambiguous word and there are 2 main parts to evolution--one part relatively random; one part decidedly less so. There are peer reviewed papers and sources that most decidedly call natural selection non-random or some synonym thereof. There are none calling it "random" or "stochastic" or "probabalistic". Mijo does not even understand what a stochastic process is and why it's not a synonym for random.

And nobody in any scientific peer reviewed paper is describing random as "of or related to a probability chart"...Nobody calls the actual probabilities of having a child with sickle cell trait random...although they may say the genes sort randomly...or stochastically...it depends what the information is that you want to convey. Only someone with a need to be unclear would call the probabilities you describe "random" nor would they say the butterfly mutation spread through the population randomly. They might say the mutation appeared at random...but natural selection ensured that it got passed on widely--not randomly.

I know you think I'm being a big meanie...but what does what I think matter. I really have had lots of experience with these types and other people online have as well and we KNOW the arguments...how the goal posts are moved.

If his goal was just to find out whether there was any evidence that evolution could NOT be described by a probability distribution...why didn't he just say that? Why does he do this semantic game where random=anything that can be described by a probability distribution and a random process = anything with any randomness in it... And stochastic=random... I'm sorry if you are missing the gymnastics he's doing so that he can sum up evolution as "scientists think this all came about randomly" but to anyone experienced with the Dover trial...this is old news. When the more someone talks, the less clarity there is, suspect woo somewhere. Or don't. I don't care. The non-woo are usually really interested in the details and facts of evolution--there is this aha moment when you understand how natural selection brings seeming design exponentially and how organisms drive eachother's evolution. But a creationist spends all their time on semantic vagaries that keep them ignorant instead of actually trying to keep abreast of the information offered.

If someone participates in this thread and cannot understand why Dawkins, Ayala, and everyone else is calling natural selection non-random, then they don't really understand how the order comes about... at least that is what Talk Origins says...and I have to agree. When you actually understand natural selection you see just how misleading and wrong it is to use that word to describe it...how obfuscating it is... and you should understand exactly why Behe et. al. want to make sure you and they never understand. What better advertisement for the misunderstanding could there be then rttjc? Scientists do not think the seeming design came about randomly. Random is the easy part of the equation...it's that which brings exponential "order" to the process that is a little harder to intuit.

Really. Ask anyone of any credibility if they think evolution is random or should be summed up that way? Ask them if they think it is misleading to call natural selection random? Ask them how they describe evolution in terms of randomness. Look for any paper anywhere that is using words the way mijo is doing to say the same thing. His explanations do not distinguish evolution from the tornado in the junkyard analogy and he shows no interest in that distinction nor in distinguishing his explanation from Behe's. Do you guys not wonder why? Ask anybody whom you consider an expert on evolution on this forum what they think of Mijo's odd questions and failure to show any interest in the answers.

Yeah, there are lots of things that would prove me wrong about Mijo being a creationist. But there is nothing that will make mijo concede that it's misleading to call evolution random and there is nothing that I could show you to prove that his argument is similar to Behe's--a known proponent of intelligent design. You guys think he's saying something, but he really truly isn't. He's using a lot of words just to restate that "scientists think this all happened by chance". Nothing in any of his words conveys the way in which order is created by the exponential survival of successful replicators selected by the environment they find themselves in. Nothing. And that is a very important concept in answering the question in his OP. If you do not understand this, you do not understand natural selection. Really.

I don't care if people agree with me or not. A lot of people have done their best to answer Mijo's question...and just like on his other threads...no answer satisfies him except the one he concluded from the beginning. Evolution is a fact...the best ways to describe it will evolve...and if anyone actually wanted the answer to the question in the OP they have it in spades. Moreover, they have all the proof they need to show that there is NOT a singular definition of random that all scientists are using and the word IS used by creationists to obfuscate understanding. Hopefully, they will start to understand how natural selection brings order to the randomness--in amazing ways... Humans have a tendency to think backwards like Von Neumann. They evolved to see design and meaning... even when it isn't there. And nature is fool of such seeming design. It's totally worth understanding the way this comes about. It can help you understand the evolution of anything. It's the same principle...the same algorithm--whatever works the best gets copied the most ad infinitum...

 

[articulett]

Essentially, yes. As you know the odds are 50% per child (assuming no screening), so the odds depend on the number of children.

If the family has 1 child the odds are 50% (1/2)
If the family has 2 children the odds are 25% (1/2)^2
If the family has 3 children the odds are 12.5% (1/2)^3

That none of the children will have sickle cell anaemia.
and so on.

How would you describe the chances of having a child with sickel cell anaemia


If only one parent was a carrier, then the odds are vanishingly that any child would have sickle cell anaemia. (The advantages of sexual reproduction and avoidance of inbreeding).

I would give the odds based on the data I had...I doubt I'd say they had a random chance. If I said it was random, I'm sure they'd presume they had the same chance as someone in the population in general having the disease.
I aim for as much clarity in my word choices as possible.

We use random in regards to mutation because mutations and changes in the DNA happen all the time whether it's good for the organism involved or not--increasing pressure in the environment does not increase mutation (with the exception of radiation) but it can alter what is and isn't selected and how fast a mutation spreads. This gives the appearance of species adapting to their environment by knowing somehow and changing accordingly. But a species stays the same species it is all it's life. The only thing that lives on is DNA... And it's the fact we don't see the failures that make the successes look so amazing or designed. We don't see the billions of aborted butterflies...but we see the results of the single success and all his progeny...

 

[mijopaalmc]

articulett-

Calling evolution "random" is no more misleading than say it can be modeled by a stochastic process.

Calling evolution "non-random" and then saying that adaptive mutations "increase the probability of survival and reproduction" is misleading.

 

[Wings]

So basically, both terms are misleading. Evolution is a complex process and it would be misleading to generalize it as either random or non-random.

Would that be an appropriate position to take on this issue?

 

[jimbob]

I have just realised that I had stated the odds for all children being carriers, but without the full sickle cell trait.

The odds of two carriers having all their children without full sickle cell are

3/4 for 1 child
9/16 for two
27/64 for three

Anyway, articulett, would you tell the parents that any child they had would have a 25% chance of having sickle cell? Then discuss the implications of this.

 

[articulett]

articulett-

Calling evolution "random" is no more misleading than say it can be modeled by a stochastic process.

Calling evolution "non-random" and then saying that adaptive mutations "increase the probability of survival and reproduction" is misleading.

Only to you and possibly another randomite or two it appears. I think that people understand Ayala and Dawkins far better than you, and I"m sure theyd find your use of words a lot more misleading. Wings read the article and wasn't confused. That's because nobody but you thinks everything that has any randomness it it can be described as a random process. Those who understand evolution, understand that it isn't the randomness that brings the illusion of the design--though creationists pretend that scientists think just that. It's natural selection--whether you understand that or can convey that understanding or not.

Everybody else seems to be able to understand what Dawkins et. al. are saying, but I'm not sure anyone understands why you are so insistent on calling evolution random.

 

[jimbob]

Aw, shucks. How can I have a debate if everyone agrees with me?

I'll try a little disagrement.

I'd say it is a nonlinear system, but with perturbations from other events weather (which I consider to be random, due to amplification of quantum effects), meteorites, volcanoes, etc...

These perturbations make the system essentially random, as well as nonlinear.

That doesn't mean that we can't model the probabilities.

If you could perform an experiment with eight identical earths all at the stage wehre humanity diverged from chimps (about 8MY ago?) and reran, to seee what would happen, humanity would not evolve, something similar might. In that time, there would be enough true randomness affecting the selection of the "winners" to stop it happening.

Identical starting conditions will not produce identical outcomes.

The only direction that evolution has is towards more optimisation for whatever the current environment is. Of course some optimisations are frequent, e.g. eyes and flight. Some are not (social, technological medium sized animals) I can't see any reason why the evolutionary niche that we occupy wasn't around 30 million years ago. It is obviously one that is fairly unlikely to be occupied. (maybe that is the answer to the Fermi paradox)...


We only think we are special because of the weak anthropic principle.

This is where VonNeumann falls down.

ETA:

Hokulele,

I would argue that over geological timeframes, which species arises is "random". As mass extinction events wipe the slate clean and those which survive are theluck ones. Which mutations prosper partially depends on which other mutations are extant at the time. The environment is not stable then, so is prone to be changed by "random" seed events (like surviving organisms' mutations).

 

[articulett]

I have just realised that I had stated the odds for all children being carriers, but without the full sickle cell trait.

The odds of two carriers having all their children without full sickle cell are

3/4 for 1 child
9/16 for two
27/64 for three

Anyway, articulett, would you tell the parents that any child they had would have a 25% chance of having sickle cell? Then discuss the implications of this.

If both are carriers I'd tell them the offspring has a 25% chance of being affected...the tests we can do and the implication of the disorder. I tell them that they have a 50% chance of having a child who is a carrier like them, but that being a carrier is not harmful--it just increases your odds of having a child with sickle cell anemia. I might tell them how carrying the trait evolved because it gave people a survival advantage in malarial prone regions--meaning that the mutation allowed their ancestors to preferentially survive.

But most scientists don't think of random as "anything related to a probability distribution"--in fact, I haven't seen one peer reviewed paper that defines randomness that way. It's too prone to misinterpretation and semantic abuse.

Is your position still that you'd answer the OP saying "there is no evidence for evolution being non-random". Do you think Mijo is more clear and descriptive than Dawkins and Ayala? Do you still think random is a good way to sum up evolution? If so, I've got to say that I just find you a little confused in a way that might not be easily remedied.

How do you distinguish the selector from the selected in regards to evolution? You seem to continually confuse the randomness of the information in the selection pool with events you think of as "random" in the selector--the environment. But evolution is driven by what gets replicated in the environment--the environment is the testing arena...and the stuff you call "random" is just part of the game...part of the elimination rounds. It would be like saying snow on the football field made the outcome random. The goal of football is to win or not win in whatever conditions present themselves. Snow doesn't make the outcome random.

You do agree that selection is biased, don't you? And you do know that bias can be seen as the opposite of random, don't you? A loaded dice is biased--not random. Why not go for clarity? And does anyone other than you think you are being more clear than Dawkins and/or Ayala and/or Talk Origins and/or the Berkeley site. Because to me, you are playing semantic games to make evolution fit mijo's vague definition of random because you think I've accused him unfairly. But put my claims aside, and answer the questions I asked.