What evidence is there for evolution being non-random?

[articulett]

Cyborg... it is a pleasure to read you...and to see it's not a communication snafu on my part. There are really only 2 randomites here...Mijo (who I'm certain is a creationist--what do you think?) and Schneibster, but as far as I can tell Schneibster is having a completely different conversation from everybody else. Jim Bob is kind of a randomite, but I can't tell what he's saying enough to understand what exactly he's claiming. I think it boils down to, he'd prefer that natural selection be called "probabilistic" rather than non-random. I think that's vague and that he still tends to confuse random components in the environment with natural selection being "random", but whatever.

I understand mijo's need to define evolution as random--he's being Behe. And Jim Bob just seems under informed and over confident in regards natural selection, creationist obfuscations, and semantic clarity. But I cannot for the life of me understand Schneibsters attachment to summing up evolution as random and his emotional investment in this word and why he thinks it's correct or clear to use language so "loosely" knowing that this is an area where creationists love to create confusion. I think that he really doesn't understand natural selection nor replicating systems...but even that doesn't make complete sense--because other "order" (the gestalt that we observe) such as landscapes and galaxies and crystal formations are imposed by the physical environment --he seems to think the "order" comes about via something else-- and then miss the further "order" in replicating systems that comes from physical environments selecting for the best "order-making and copying machines"

But he became so insulting, that I have begun to think it amusing that he is being taken in by Mijo. Though I am not sure any of them are actually understanding each other. I truly think you are being as clear as humanly possible...but I also thought Dawkins and Ayala were clear. But I think the bottom line is that no amount of evidence will make some randomites understand that they are being vague and uninformative...and that, despite their assertions to the contrary, most scientists would assert that natural selection is the answer to the OP question.

Why would someone have a semantic interest in whether scientists use the term "non-random" in describing natural selection? Why would someone deny all evidence that some parts of evolution might more clearly be described as "not random"-- especially when there are two parts to evolution--one part more random (though not truly random by the strictest definition of the word) than the other (which selects from the pool of randomness to bring incremental "order")?

Why wouldn't any scientific person understand this usage of the word after seeing example after example of creationist obfuscation using that exact word? Why wouldn't someone aim for the clearest and most useful means of describing evolution and would any credible scientist actually answer the question in the OP by saying, "there is no evidence for evolution being non-random"? They might say...it's a vague question...or what do you mean by random?...or what is it you are really asking? But I am very certain that most would answer as Dawkins did in regards to Behe's similar claim...or as I answered...as you answered...as Ivor answered...as Ayala answered...as Talk Origins asserted...and as the Berkeley site said--I mean, what a panapoly of answers one chooses to ignore in asserting that it's meaningful to call evolution a random process. Why would anyone be so vague after seeing a thread of this length testifying to the misunderstandings and ambiguity associated with the term? Anyone, other than I creationist, I mean?

 

[articulett]

I'd say if the goal is to explain evolution simply and clearly, then describe like those who have explained it simply and clearly to many people. Moreover, aim not to sound like Behe who has obfuscated understanding to many people. Should clarity be your goal, I strongly recommend you don't use words like probabilistic and random when talking about natural selection. Most biologists consider this the most important part of understanding how the illusion of design comes about. The words they use tend to be "deterministic", "non-random", preferential survival, culling, elimination rounds, etc. They tend to use words that distinguish the randomness of mutation from the "force" or "engine" that physical constraints in the environment impose upon the information carriers which brings "order" to the system. That which we can observe and measure in regards to evolution is order imposed by the environment selecting the best information replicators within it. The information itself is what is changing (relatively) randomly.

It's very simple. I can't imagine why anyone would want to use ambiguous wording. I, personally, wouldn't use the word "probability" unless I was talking about a specific probability, and then I would give that specific probability. Otherwise, people will make the error that Jim-Bob is making and confuse events in the environment that they find "unpredictable" with this muddled notion that natural selection IS "random".

If you define everything that contains randomness as "random" or "probabilistic"--then you've made the language so loose, that it is virtually useless for conveying understanding. If your goal is clarity--don't use language like the unclear. There are lots of explanations on this thread and on Talk Origins and all over the web. Who is the clearest? Who conveys the information to the most? Who conveys the most information in the fewest words? I don't think anyone in the world would conclude it's Behe. Evolution is a theory, fact, and principle like gravity. And, like gravity, if someone wanted to know what the "non-random" aspects were... most scientists would have a lot more to say than "there are none".

 

[mijopaalmc]

articulett-

How is natural selection not based on a probability distribution?

When a subpopulation that reproduces is selected from the population at large, a sample of the population at large is taken. Sampling is an inherently statistical process, meaning that the proportion of individuals with a given set of genes that appears in the population that reproduces may be larger or smaller that the "true" proportion in that population at large. In fact, it is most likely that way.

If the proportion if individuals with a given set of genes varies in a non-statistically significant way from the proportion of individuals in the population at large, there is no selection either for or against that set of genes. If, however, the proportion if individuals with a given set of genes varies in a statistically significant way from the proportion of individuals in the population at large, there is selection either for or against that set of genes, depending on whether the proportion is significantly larger or significantly smaller than in the population at large.

What is so confusing or obfuscating about that description, besides the fact that you obviously don't want to understand and will go to great length to avoid doing so?

 

[jimbob]

Cyborg... it is a pleasure to read you...and to see it's not a communication snafu on my part. There are really only 2 randomites here...Mijo (who I'm certain is a creationist--what do you think?) and Schneibster, but as far as I can tell Schneibster is having a completely different conversation from everybody else. Jim Bob is kind of a randomite, but I can't tell what he's saying enough to understand what exactly he's claiming. I think it boils down to, he'd prefer that natural selection be called "probabilistic" rather than non-random. I think that's vague and that he still tends to confuse random components in the environment with natural selection being "random", but whatever.

In a stable environment, over thousands of generations, the probabilities are such that "nonrandom" would work as a starting point. You are at the tail of the distribution, and are dealing with large numbers.

Assuming a beneficial mutation occurs in one individual, then it takes many generations before that mutation is well enough established in the population to be fairly immune to "everyday" "random" events. (Where a particular storm hits, whether a predator was in the "wrong" place etc...)

Until the mutation has escaped the population bottleneck, it is very vulnerable to "chance" events. A mutation causing a similar trait might occur later or not, it depends if the ecological niche is (still) available; a small carnivore is not going to evolve into a large carnivore if there already is a large carnivore in that environment.

Over geological timeframes, the environmnet is altered drastically by events that are certainly unpredictible, and which cull large numbers of species, and the culling is essentially random.

The peripheral effects of any meteor strike are unpredictible, and I would argue random, being affected by chaotic systems.

I understand mijo's need to define evolution as random--he's being Behe. And Jim Bob just seems under informed and over confident in regards natural selection, creationist obfuscations, and semantic clarity. But I cannot for the life of me understand Schneibsters attachment to summing up evolution as random and his emotional investment in this word and why he thinks it's correct or clear to use language so "loosely" knowing that this is an area where creationists love to create confusion. I think that he really doesn't understand natural selection nor replicating systems...but even that doesn't make complete sense--because other "order" (the gestalt that we observe) such as landscapes and galaxies and crystal formations are imposed by the physical environment --he seems to think the "order" comes about via something else-- and then miss the further "order" in replicating systems that comes from physical environments selecting for the best "order-making and copying machines"

How is mijo being Behe?

He did ask about gaps in the fossil record, but now accepts that this is illusionary.

He rejects "irriducible complexity" as an argument, and states that humanity evolved without any outside agency or designer.

Mijo, please correct me if I have misrepresented your previous posts in this thread.

But he became so insulting, that I have begun to think it amusing that he is being taken in by Mijo. Though I am not sure any of them are actually understanding each other. I truly think you are being as clear as humanly possible...but I also thought Dawkins and Ayala were clear. But I think the bottom line is that no amount of evidence will make some randomites understand that they are being vague and uninformative...and that, despite their assertions to the contrary, most scientists would assert that natural selection is the answer to the OP question.

But Dawkins implicitly and explicitly treats selection as probabilistic.

Though I am not sure any of them are actually understanding each other.

I understand mijo, he has agreed with me, as has Schneibester. We have slightly different emphases but to any numerate person, it is easy to see that we are agreeing. I also suspect that none of us are theists.

You were the person saying that you had toruble in discussions with creationists and IDers. Maybe it is because there is little to separate your statements from some of the more secular (C of E) clergy, and their attitude to evolution.

"Yes we evolved, but the laws of evolution were such that our existance was inevitible, and an omniscient being would know that. God just set up the right conditions, and created us using evolution."

I have heard that argument many times since school.

"All God needed to do was select which organisms survived"

Especially if you state that aircraft (design) evolved, as both you and cyborg have.

"Just as aircraft designs evolved, and they had many designers; so we evolved and had one designer."

Can't you see that that statement is playing into the hands of any ID proponent.

 

[jimbob]

http://www.talkorigins.org/faqs/chance/chance.html

Natural selection is not random: it is the determinate result of sorting processes according to relative fitness. It is stochastic, in the sense that better engineered features can fail for reasons of probability (they may meet accidents unrelated to their fitness), but that poses no greater threat to the scientific nature of evolution than it does for, say, subatomic physics or information theory.

There are scientists and philosophers who think that probabilities represent a real indeterminacy in the world; that even if you had, in principle, full information about all causes for a system, it would still be possible only to predict the distribution curve rather than the outcome for any single object. This is called the propensity interpretation (Beatty and Finsen in Ruse 1989), and holds that real things have a real propensity to behave in a range of ways rather than a real set of properties that will specify a strict determined outcome. Whether this is true or not is not relevant to evolution as such, for if it is true, then it is true of everything, and not just living things.

It is stochastic, in the sense that better engineered features can fail for reasons of probability (they may meet accidents unrelated to their fitness), but that poses no greater threat to the scientific nature of evolution than it does for, say, subatomic physics or information theory.

This is what I have been saying. "It is stochastic, in the sense that better engineered features can fail for reasons of probability (they may meet accidents unrelated to their fitness),"

And you can assess this using probabilities.

 

[mijopaalmc]

I would say that it is Dawkins and Ayala who are being vague and uninformative. They are the ones who are describing evolution in two contradictory ways in so far as "random" means "[o]f or relating to a type of circumstance or event that is described by a probability distribution" and they are describing evolution as both "non-random" and "non-random" and as "increas[ing] the probability of survival and reproduction of their carriers at the expense of maladaptive, or less adaptive, variations". Furthermore, if Danwkin and Ayala's beliefs mirror articulett's beliefs (as articulett like to claim Dawkins and Ayala do), Dawkins and Ayala believe that the definition of "random" that I have used throughout this thread is meaningless. This is an absurd and intellectually dishonest stance as the fields of probability theory and statistics are firmly based on that definition, and I don't think Dawkins or Ayala, or even articulett, would claim that probability theory and statistics are meaningless as fields of study in pure and applied mathematics.

 

[jimbob]

mijo, from the extended phenotype:

"A selection pressure as weak as 1 in 1000 would take only a few thousand generations to push an initially rare mutation to fixation".

"If the selection pressure we are discussing is very strong, that is if one replicator makes its posessors very much more likely to survive than its alleles do "

Dawkins invokes a probabilistic treatment, as he would have to, given his background.

Articulett, are you sure that your statements about emotional invsestment couldn't apply more to you than me?

If mijo is a creationist, he has chosen a bloody odd way to proclaim his faith, by attacking it with rational argument.

 

[articulett]

articulett-


What is so confusing or obfuscating about that description, besides the fact that you obviously don't want to understand and will go to great length to avoid doing so?

Actually, I think I am one of the few people who went to great lengths to try and understand you. I don't think anybody understands what you are saying or could sum it up any better than they could sum up Kleinman or Behe. I just think you are using a lot of words and semantic tricks to justify calling evolution random. And, as cyborg said, that's fine. I'm glad it makes you happy, but I don't engage in discussions with creationists because I do not find them amenable to basic logic, and I have no interest in semantic games.

 

[steenkh]

In a stable environment, over thousands of generations, the probabilities are such that "nonrandom" would work as a starting point. You are at the tail of the distribution, and are dealing with large numbers.

That is how I see it (although I do not know what "tail of the distribution means: I think it sounds ominous).

Assuming a beneficial mutation occurs in one individual, then it takes many generations before that mutation is well enough established in the population to be fairly immune to "everyday" "random" events. (Where a particular storm hits, whether a predator was in the "wrong" place etc...)

But surely, over longer time scales, this random effect is cancelled out?

Until the mutation has escaped the population bottleneck, it is very vulnerable to "chance" events. A mutation causing a similar trait might occur later or not, it depends if the ecological niche is (still) available; a small carnivore is not going to evolve into a large carnivore if there already is a large carnivore in that environment.

Yes.

Over geological timeframes, the environmnet is altered drastically by events that are certainly unpredictible, and which cull large numbers of species, and the culling is essentially random.

I would have thought that over geological timescales, only gigantic events would still be regarded as random. Everything else would have been reduced to a "noise level". Apart from these events evolution would be regarded as non-random, in the sense that it would be obviously random which solutions will be chosen for the survival of the various species, and obviously random which species survive, and which become extinct, but on the other hand it would be obviously non-random that more and more ecological niche will be filled, and that complexity grows. The tornadoes are busy churning out more and more complex air planes from the scrapyards.

The peripheral effects of any meteor strike are unpredictible, and I would argue random, being affected by chaotic systems.

Yes, but what have isolated random events to do with the principle of evolution? They may alter the course of evolution, but not change the way evolution works, and the non-random aspects of evolution will ensure that after enough time, all the niches will be filled out again.

 

[articulett]

http://www.talkorigins.org/faqs/chance/chance.html





This is what I have been saying. "It is stochastic, in the sense that better engineered features can fail for reasons of probability (they may meet accidents unrelated to their fitness),"

And you can assess this using probabilities.

Yes, but the very first sentence is that natural selection is not random, but deterministic. Wasn't that mijo's question? It's "stochastic" in a sense...but for mijo, stochastic is a synonym for random as is anything to do with a probability description. If the three of you (Mijo, Schneibster, and you) think that this means that there is no evidence for evolution being not-random or that it makes sense to sum up evolution in the weird way you seem to be doing it, then so be it. I am saying that it is no more clear than Behe. That your descriptions sound like you don't actually get natural selection...that you think it's informative in some way to call evolution random or natural selection "probabilistic".

Lots of words can be used. But no one actually conveying information to people is saying "evolution IS random" or "evolution is probabilistic" or "evolution can be summed up as random mutation and probabilistic selection". If you cannot understand why nobody is doing that. Or why no actual person who understands or teaches evolution would use language in this way or why it's misleading and uninformative after all this--then I don't think anything I can say will help. But that is exactly like Behe. It truly is. He's not really "wrong"--but he's so misleading and obfuscating that, well, he comes off as "wrong"--or ignorant--or misinformed...or unclear on what natural selection is.

I take it that for the three of you the answer to Mijo's OP question is that there is no evidence (that will satisfy you) that evolution or any part of it is non-random, right? I maintain that is Behe's position.

And my only emotional investment is that people understand what natural selection is and that they aren't made ignorant by the wedge strategy of obfuscation. I don't care what words are used. I am quite certain that the best ways of conveying the concept will evolve. Everyone who writes on the topic is clear to me except Mijo, Schneibster, and you--and as far as I can tell you are not even making the same argument and seem clueless as to how your misunderstanding is identical to Behe's misunderstanding even as you illustrate what a clever obfuscation technique it is.

You all sound very muddled on a very simple concept because you have some sort of need to use the word random or words you feel are synonymous with it-- that is why no text or peer reviewed article is saying things the way you are saying them and that is why you need to twist and play semantic and definition games to claim they are. As far as I can tell, the majority of people would find you, Mijo's, Schneibsters, and Behe's explanations of evolution equally uninformative and muddled. And the majority would find Cyborg, Dawkins, Ayala, Talk Origins, Sphendisc, Ivor, Paul A. and Berekely far more informative and easy to understand and "correct". If your goal is clarity, I suggest any of the above. If your goal is convincing yourself that there is no evidence that evolution is "non random" or that Dawkins et. al. is wrong and your are right--then you've achieved that. I just don't think anyone else shares that view. The questions you ask are almost as leading and loaded and useless as the questions mijo asks. Unless you wanted to imply an answer in the first place.

And jim-bob...I and many others who are actually familiar with creationist obfuscation techniques have answered your question regarding playing into the hands of creationist, but you just can't hear it. Suffice to say, that most biologists consider understanding natural selection (how an environment...including the "random" stuff in it affects the ways life forms evolve) is THE KEY to understanding evolution. Your explanation muddies that understanding. As does Behes, Mijos, and Schneibsters. For some people, that is intentional even if not admitted.

 

[articulett]

I would say that it is Dawkins and Ayala who are being vague and uninformative. They are the ones who are describing evolution in two contradictory ways in so far as "random" means "[o]f or relating to a type of circumstance or event that is described by a probability distribution" and they are describing evolution as both "non-random" and "non-random" and as "increas[ing] the probability of survival and reproduction of their carriers at the expense of maladaptive, or less adaptive, variations". Furthermore, if Danwkin and Ayala's beliefs mirror articulett's beliefs (as articulett like to claim Dawkins and Ayala do), Dawkins and Ayala believe that the definition of "random" that I have used throughout this thread is meaningless. This is an absurd and intellectually dishonest stance as the fields of probability theory and statistics are firmly based on that definition, and I don't think Dawkins or Ayala, or even articulett, would claim that probability theory and statistics are meaningless as fields of study in pure and applied mathematics.

I see, so Dawkins, Ayala, etc. are being uninformative, while you, mijo are being "informative" even though you have provided not a single peer reviewed paper that defines "random" as you do and I have provided multiple sources including a peer reviewed paper that has defined it differently. Everyone's point is that random is an ambiguous word and it seems to confuse understanding when used in reference to natural selection. Claiming that everyone agrees with your definition and description is not the same as having a peer review source saying random is "relating to or described by a probability distribution" or "random is a synonym for stochastic".

 

[articulett]

Jim-bob your confusion comes from your not understanding that life evolved to live on this planet which contains elements that you describe as random. You are muddling things to highlight such event. A sneeze could save an organisms life by getting rid of a pathogen or it could give it's location away to prey--but these assorted probabilities or "randomness" are PART of natural selection--they are the rules of the game. The DNA that gets itself copied in the most by making the best copiers in the environment it's in are the KEY to evolution...natural selection. You just seem to miss this and get muddled over it and think you understand something or that Dawkins or others are saying what you are saying, but they are just not. This conversation is happening over your head. Nobody denies probabilities. It's just that you can't see the forest for the trees. You can't get the basic principle because you are too caught up on what you think of as random.

And you truly are a beautiful illustration of exactly what creationists hope for. If you obfuscate natural selection enough people will believe that scientists think that all this complexity and seeming design just happened randomly. But we DO NOT THINK that. We understand how it happened and why it looks designed. And it may be too late for you to understand the same. And for me, that is something that makes me a little sad--yes. I feel like logic is the best tool we have to fight ignorance, but it clearly doesn't work on people certain that they have the "right answer".

You aren't "wrong"--just muddled and missing some key points. It reminds me of my son trying to learn algebra and getting mad that "x" could be any number--he wanted to know what number it was! I feel that you are stuck like that...but in a more sophisticated way-- yet you are still "not getting it"--not understanding why scientists would say that natural selection is not random or be careful not to confuse what you term "random events" with natural selection. Life evolved to live on a planet that has the random events you speak of. All physical events are part of the "elimination rounds" directing the evolution of life.

 

[mijopaalmc]

articulett-

What don't you understand about the two sentences "evolution/natural selection is not random" and "evolution/natural selection increases the probability of of a well-adapted organism's survival" being contradictory statements?

What don't you understand about a stochastic process being defined by probability and a random variables?

 

[articulett]

It's just so backwards and egotistical...
"the world is perfectlly made to bring forth humans and all the wonderous life we see around us."

No. All the life we and all the life we see around us evolved because it could in the environment it found itself it. Each ancestor was successfully able to copy it's info. including all the info. that helped it develop a good copying machine. It did it because it could. In the course of doing whatever it was programmed to do via it's DNA, it got itself copied.

Do you not see what Cyborg is telling you when he says you have it backwards? Whenever you define fitness you are doing just that same kind of backwards thinking. You are using different frameworks for words like fit and probable and random and all these words are words that are relative to something else. Fit compared to what? Probable compared to what? Random compared to what? Evolving from what? Genotypes evolving or phenotypes evolving? Evolution of what?

As for the basic principle of evolution, you randomites just sound as muddled as Behe and the selected quote that Dawkins bases his review on. You sound as if scientists would actually say "evolution happened randomly". That is so simplistic and misleading that no scientist I know would ever say such a trite thing unless they were using it colloquially to mean "unplanned" or without thought. It's just vague in the exact way creationists are vague and obfuscating. And I can't understand why anyone would claim that it's not or insist on being that muddled sounding. If you agree with Behes statement about evolution being based on randomness, then most intelligent people will respond as Dawkins and Ayala and Talk Origins and the Berkeley site have...and the majority of people on this thread. Why because it's muddled and backwards and misleading and uninformative and vague and too easy to confuse people with.

I just want to make sure that the people who are actually interested in understanding natural selection, get a chance to do so. I want to make sure that those who want to know the answer to the question in the OP--understand exactly why Dawkins et. al. say natural selection is not random. I want others to understand exactly what Behe is doing and why.

Natural selection is about everything in the environment--including the events and hypotheticals you want to call random. You make the word random meaningless because your definition makes ALL environments random. You focus on the wrong part of the equation and miss the main point completely.

You just have it so backwards.

 

[articulett]

articulett-

What don't you understand about the two sentences "evolution/natural selection is not random" and "evolution/natural selection increases the probability of of a well-adapted organism's survival" being contradictory statements?

What don't you understand about a stochastic process being defined by probability and a random variables?

Nobody but you uses random to mean "probability" and random components do not make a process random.

http://scienceblogs.com/pharyngula/...sts.php?utm_source=mostactive&utm_medium=link
I think only you (and possibly some creationists and maybe another randomite) would actually conclude those two statements are contradictory. But then no one else defines random quite as loosely as you do despite your claims that everyone does. Oh...and just because probability and random variables are involved...doesn't mean the thing being described IS stochastic. This is logic 101. Just because stochastic involves probabilities, doesn't mean that all things involving probabilities are stochastic. Really, you should brush up on logic. Also, clarity is of essence in conveying understanding...so you may want to lay off the semantic vagaries unless your goal is to purposely not understand...and use definition the way peer reviewed sources are using them. Just saying...

Also, the last sentence is backwards...natural selection doesn't increase a well adapted organisms survival, rather, a well adapted organism is less likely to be culled by natural selection. But even then...adapted is in direct reference to the environment--the body and surroundings it finds itself in. You are just so muddled in your explanation. The environment (including anything that you would call "random") is what chooses the best reproducers from the pool of randomness. It's so simple...but you just can't get it. You ignore any and all information that would allow the synapses to connect and understanding to occur. THE ENVIRONMENT IS THE SIEVE THROUGH WHICH ONLY THE MOST SUCCESSFUL REPLICATORS PASS.

Any information which confuses that fact, muddles understanding. It is not random nor is it stochastic that some things pass through a sieve and others do not. It is not random that some things survive successive elimination rounds and others do not. And everything you say is akin to saying "it is so random" or "It's random that the most successful replicators get passed on". Because everything you say can be boiled down to this vagary, it's a useless and uninformative definition that only creationists and the ignorant will find useful. Dawkins isn't contradictory--your definition is just so muddled that you can't see how ignorant and clueless you sound. I doubt you could successfully convey a thing about evolution to anyone.

Math problems are often defined by probabilities and random variables...it doesn't mean that it makes sense to call these "random math problems" or "stochastic processes". In fact, I'm sure it would make much more sense to say, "the equation for a slope is not random" despite the fact that it contains "randomness". The same is true of evolution...PARTICULARLY natural selection. As you can see, it's a semantic argument...which is why most scientists avoid using terms like random without defining their meaning except when dealing with semantic obfuscation by those who make natural selection difficult to understand by summing up evolution as random. I imagine most mathematicians wouldn't have to convey the notion that the equation for a slope is not random because there aren't a bunch of blow hards trying to confuse people by claiming it is.

 

[Schneibster]

I understand mijo, he has agreed with me, as has Schneibester. We have slightly different emphases but to any numerate person, it is easy to see that we are agreeing. I also suspect that none of us are theists.

You suspect correctly in my case; I am an atheist (and not a nice one). I suspect that the bolded part of your quote is the problem in articulett's case.

Can't you see that that statement is playing into the hands of any ID proponent.

Never mind; I've been trying to tell her that since my entry into the thread. I think you're wasting your time.

 

[Meadmaker]

Ok, I've been out of this one for a while, but I have a question for the randomites. Do you believe what happens next is constrained completely by all of what happened before?

I'm guessing the answer will be "no". If so, how much wiggle room do you think there is for the outcome of the next event, given all of the events up to that point?

How would you tell the difference between x+1 = f(x) + err, where err is a truly random component and x+1 = f(x) + g(x-1, x-2, ..., x-N), where g() is a non-linear function from hell over all events till the present?

There's no way to tell the difference.

From what I have read, the general belief among physicists is that quantum events are truly random. Also, the general belief is that the truly random events at the quantum level can have effects that are measureable at the macroscopic level, which would imply that the macroscopic level events are truly random as well.

A famous example would be the distribution of galaxies in the universe. One of the great cosmological problems is figuring out why the distribution of mass in the universe is non-uniform. Current belief is that in the moments following the big bang, quantum fluctuations within the tiny, tiny, universe made things slightly non-uniform, and the result was clusters of galaxies.

Similarly, why do we have a universe consisting almost exclusively of matter, with almost no anti-matter? The general belief is that it could have just as easily been the other way, but some quantum fluctuation caused there to be a slight imbalance in favor of matter, and the anti-matter was almost entirely annihilated as a consequence. In other words, matter won the coin toss, and the coin toss was truly random.

Is that view wrong? Could it be that there were actually hidden variables that made those quantum events inevitable, producing a matter-based universe with a galaxy distribution preordained by the initial state? Sure, that could be the case, but there would be no way to tell. Einstein famously said that God did not play with dice. He might be right, but most physicists think he was wrong.

Of course, that isn't really the argument going on in this thread. Most of the "random" events being discussed in this thread are not quantum based, at least not at a proximate time to the event. Once upon a time, I gave an example of a bird with a beneficial mutation being eaten by a snake while the bird was still in its egg. That isn't a "truly" random event the way you described. That's a nonlinear function from hell event. The argument in this thread doesn't really distinguish the two sorts of random events, and just calls both of them random.

 

[Schneibster]

There's no way to tell the difference.

That's the whole point, you can't distinguish between these.

From what I have read, the general belief among physicists is that quantum events are truly random. Also, the general belief is that the truly random events at the quantum level can have effects that are measureable at the macroscopic level, which would imply that the macroscopic level events are truly random as well.

<snip>

Is that view wrong? Could it be that there were actually hidden variables that made those quantum events inevitable, producing a matter-based universe with a galaxy distribution preordained by the initial state? Sure, that could be the case, but there would be no way to tell. Einstein famously said that God did not play with dice. He might be right, but most physicists think he was wrong.

Actually, some recent experiments have managed to rule out "local hidden variable" theories, although there is some equivocation regarding loopholes. By and large, however, based on the available evidence, which is quite good but not yet absolutely conclusive, you're correct, most physicists seem to believe that local hidden variables have been essentially ruled out. Global hidden variables, which is part of what you're talking about here, are basically unfalsifiable at this time; hypotheses about them are untestable, and so far have mostly the status of what are called "interpretations" of quantum mechanics. No one sees a way to test among the interpretations; some genius may come along and come up with a way, but the barriers are extremely high. It is therefore safe to say at this time that global hidden variable theories are indistinguishable both from one another, and from physics without them; and as such, are not in the province of physics proper.

To put this another way, if quantum events are not at least some of the time random or acausal, there is no way we could ever show it; for things to work the way they do, but quantum events not to be truly random, the non-random aspects would have to be so thoroughly hidden that we cannot see a way to measure them, and they therefore effectively do not exist as far as measurable physical effects (i.e. science ) are concerned. Most physicists therefore feel that Occam's Razor makes random quantum behavior the most likely explanation.

Of course, that isn't really the argument going on in this thread. Most of the "random" events being discussed in this thread are not quantum based, at least not at a proximate time to the event. Once upon a time, I gave an example of a bird with a beneficial mutation being eaten by a snake while the bird was still in its egg. That isn't a "truly" random event the way you described. That's a nonlinear function from hell event. The argument in this thread doesn't really distinguish the two sorts of random events, and just calls both of them random.

This appears to me to have much the same character as the physics question about randomness at the quantum level and global hidden variable theories; it is, in other words, a distinction without a demonstrable difference.

In other words, the "non-randomites" are arguing about philosophy, not science.

 

[mijopaalmc]

In other words, the "non-randomites" are arguing about philosophy, not science.

I have always thought that the non-randomites' argument rested more on what we don't know than what we don't know, because they seemed to be arguing "just because evolution appears random doesn't mean it is". However, since it is impossible, at least from our current knowledge, to predict every interaction that an individual will have with its environment and other individuals in its environment (which is what articulett claims is the basis for natural selection), it is impossible to tell if a particular individual will be selected over another individual, especially if they have identical or nearly identical sets of genes that contribute to their phenotypes. Thus, given our best current empirical evidence bother from the field and from modeling, it is perfectly acceptable to say that evolution is a stochastic process, even though a completely deterministic process may be discovered at some future date. Nevertheless, making the statement "evolution is non-random" is completely unsupportable given the current empirical evidence.

 

[jimbob]

That is how I see it (although I do not know what "tail of the distribution means: I think it sounds ominous).

If you look at the Bell curve, the tails are the outliers, far from the vast majority. ^WP
say seven or eight standard deviations, and the decay in liklyhood is almost

But surely, over longer time scales, this random effect is cancelled out?

In an environment with stable selection pressures, yes, or at least one can (in principle) calculate the probabilities that this will be the case. In other words one can say with how much certainty you can treat it as nonrandom as a simplification....
<snip>

I would have thought that over geological timescales, only gigantic events would still be regarded as random. Everything else would have been reduced to a "noise level". Apart from these events evolution would be regarded as non-random, in the sense that it would be obviously random which solutions will be chosen for the survival of the various species, and obviously random which species survive, and which become extinct, but on the other hand it would be obviously non-random that more and more ecological niche will be filled, and that complexity grows. The tornadoes are busy churning out more and more complex air planes from the scrapyards.

If species tend to start from small initial numbers of populations acquiring a beneficial mutation, then the initial beginning of speciation will involve small numbers of individuals, and thus at the start be susceptible to chance events. The ecosystem, and thus the environment for the other (competing, prey, and predator) organisms will depend on whether this mutation propagates.

I agree entirely with your point about ecological niches, which I regard convergent evolution as the corollory.

Physics Metaphor in the spoiler:

In semiconductors, there are "traps" which are due to crystal imperfections, and which lie at intermediate energy levels from the main energy bands.

Because these lie at different energies, the chances of an electron being thermally excited to reach these traps depends on the temperature ans is probabilistic according to fermi-dirac statistics.

An empty low energy trap has a far higher chance of being filled compared to higher energy traps.

I sometimes think of the ecological niches in a similar way, the niche for a "large herbivore" (and consequently "large carnivore") will be at a lower energy than that for a "technological social animal". The niche might still be there, but one is far more likely to have been filled.

Would the niche for "technological social animal" have been avialable 100 million years ago? I would argue that it probably would have been, but wasn't filled.

Maybe one needed large mammals, so try 35 million years ago.

We are no more complex than the animals around then (possibly except for our brains). Would a "medium sized technological social animal" have been able to thrive had it existed; probably, but the niche wasn't filled, and now it is. I would argue that this niche is filled now, not just because the conditions were right, but that the dice came up right as well this time...

I would say a probabilistic treatment of evolution leaves very little room for a creator, as the only way of ensuring that humans evolved would have been to interfere in the "free will" of many organisms, which is anathema to many deists. And no, I don't want to get into that discussion here...

Yes, but what have isolated random events to do with the principle of evolution? They may alter the course of evolution, but not change the way evolution works, and the non-random aspects of evolution will ensure that after enough time, all the niches will be filled out again.

I am arguing that to quantify the chances of various evolutionary outcomes, one needs a probabilistic treatrment, this might tell you that a non-probabilistic treatment is a reasonable simplification to a 99.x% probability

The point about natural selection is not that the "fittest" reproduce, but that a mutation increases the chances of an organism reproducing. This can be quantified.

OK, I am a physicist, "to quantify something is to know it". (Not always true, but if one approach can give you useful numbers, and another can't then the quantifiable approach is superior to me.) If the non-numeric approach actually stops you assessing the numbers it is misleading.