What evidence is there for evolution being non-random?

[mijopaalmc]

But I do.

And irony always puts a little spring in my step. I'm also fond of the whooshing sound I imagine as I see it flying over his head.

You know, it is ironic that you think I don't get what you say.

You seem to miss that it is the statistics that matter more that the rules, because no matter how well the community of replicators (that is your cherished terminology, right?) plays the game in side its vehicle, there is no guarantee that the same community of replicators inside a different vehicle will fare identically. Thus it is possible for you to have one community of replicators survive while and identical community of replicators doesn't. In short, it is not the cards that are dealt (the replicators themselves) that change it is the rules of the game (the events and entities that are external to the replicators) that change for each community of replicators.

 

[steenkh]

I think you're saying that if a gene can't get into a replicator or can't form a replicator it is automatically "unselected" and unfit. I think everyone agrees on that.

I do not think that fitness can be an attribute of a single gene (unless that gene actually kills its owner). A gene that makes an individual blind can be devastating in some instances, but can be neutral in other circumstances, like when the individual lives in a cave or has some other means of compensating for the loss of sight. Fitness must be a description of the whole, not the individual genes.

And I also do not think that "unselection" of a gene is immediate. An individual fish might survive with a larger size than the mesh in the fishing nets, but given time, the unselection process will weed it out because the pressure of the fishing boats will remove the larger fish.

I see the point of the random events that have little to do with the normal unselection process, and I agree that this should be treated as noise that will be evened out over time, making the unselection process deterministic.

The creationists I have encountered are more interested in the end product than in the process, and they try to argue that it is highly unlikely that a random process would lead from a single-celled creatures to humans. To counter this argument it is immaterial whether evolution is random or not, because even if evolution is purely random, it is obvious that the probability is larger than zero, and we argue that it has happened at least once. The creationists then often ask if I really believe that a toad could evolve into a prince if humans had become extinct. My answer here is that a toad could conceivably evolve into something intelligent, but it would surely not be humans. Toads and humans split long ago on the tree of evolution, and toads cannot become humans because evolution never backs down a branch but always continues in the direction it has taken.

I mention this line of argumentation because it reminds me of the argument about the tornado that supposedly cannot form a 747 from a junkyard. If we ignore the fact that evolution only makes actions that are not impossible, whereas the 747 needs actions like soldering that could never be performed by a tornado, I think we can argue that the non-random aspect of evolution ensures that something will come out of our evolutionary tornado, but the random aspect makes it impossible to know if it will be a 747 or a luxury liner.

 

[articulett]

Agreed across the board steenkh--and you said it said better than I could.

Plain old randomness is the beginning of many complexities--a selection force speeds and biases the result.

Organisms that have information that help them survive and reproduce pass on more of the information to survive and reproduce. That which is selected favors more such selection.

You sound like you understand it better than mijo and Jim Bob to me.

 

[articulett]

You know, it is ironic that you think I don't get what you say.

You seem to miss that it is the statistics that matter more that the rules, because no matter how well the community of replicators (that is your cherished terminology, right?) plays the game in side its vehicle, there is no guarantee that the same community of replicators inside a different vehicle will fare identically. Thus it is possible for you to have one community of replicators survive while and identical community of replicators doesn't. In short, it is not the cards that are dealt (the replicators themselves) that change it is the rules of the game (the events and entities that are external to the replicators) that change for each community of replicators.

Ah yes...yet another hypothetical diversion so you can once again claim that evolution is random. It is mijo. By your definition, evolution is random. As for the rest of the scientific community, all of those things that are external to the replicators--we call those, "natural selective forces". We are assuming these replicators live in the natural world, right? Hence they are biasing what is and isn't selected via (gasp) NATURAL SELECTION. The replicator has to play the cards it's dealt in the environment it's in amongst the other players in the game to "win". (See cyborgs Poker analogy for a clue). Having a good hand, doesn't guarantee success. And some can win with a bad hand. But you want to say the winners are chosen through a random process.

Misleading.

Now what was your goal again?

 

[articulett]

Oh my--Dawkins said it again--and in a review of Behe's book:

http://richarddawkins.net/article,1360,Inferior-Design,Richard-Dawkins

The crucial passage in "The Edge of Evolution" is this: "By far the most critical aspect of Darwin's multifaceted theory is the role of random mutation. Almost all of what is novel and important in Darwinian thought is concentrated in this third concept."

What a bizarre thing to say! Leave aside the history: unacquainted with genetics, Darwin set no store by randomness. New variants might arise at random, or they might be acquired characteristics induced by food, for all Darwin knew. Far more important for Darwin was the nonrandom process whereby some survived but others perished. Natural selection is arguably the most momentous idea ever to occur to a human mind, because it — alone as far as we know — explains the elegant illusion of design that pervades the living kingdoms and explains, in passing, us. Whatever else it is, natural selection is not a "modest" idea, nor is descent with modification.

The review is excellent...and you will learn everything you need to know about abuse of the word random by creationist. Plus the Dogs from Wolves analogy is a pleasure for any non-creationist to read as is most anything Dawkins writes is if you are a person interested in understanding evolution.

Oh...and it ends with this suggestion

If correct, Behe's calculations would at a stroke confound generations of mathematical geneticists, who have repeatedly shown that evolutionary rates are not limited by mutation. Single-handedly, Behe is taking on Ronald Fisher, Sewall Wright, J. B. S. Haldane, Theodosius Dobzhansky, Richard Lewontin, John Maynard Smith and hundreds of their talented co-workers and intellectual descendants. Notwithstanding the inconvenient existence of dogs, cabbages and pouter pigeons, the entire corpus of mathematical genetics, from 1930 to today, is flat wrong. Michael Behe, the disowned biochemist of Lehigh University, is the only one who has done his sums right. You think?

The best way to find out is for Behe to submit a mathematical paper to The Journal of Theoretical Biology, say, or The American Naturalist, whose editors would send it to qualified referees. They might liken Behe's error to the belief that you can't win a game of cards unless you have a perfect hand.

(Irony?)

And Mijo--you should do the same. Really. (Damn, I wish I could write like Dawkins.)

In fact, randomites, Dawkins actually had the temerity to call natural selection "nonrandom"--I even bolded it for you!--Write to your favorite peer review board and let them know that he's wrong, wrong, wrong and that you and Behe have the probability distributions to show why this is so. Dawkins actually called natural selection a nonrandom process. It's right there...on his website...in a critique of Behe's book--Behe, the number one creationist blatherer.

It's what I've been saying. All of it. The term nonrandom is used because of creationist obfuscation. And, just like Dawkins, in PNAS peer reviewed articles it is used in reference to Natural Selection.

No need to insult me or tell me I'm wrong or how great your model is. I was just trying to fill you in on the parts of evolution that scientists consider "nonrandom" per the OP question--you know--why it's misleading when you describe a two part model with the same nebulous word. (I'll pretend I'm getting apologies for those who thought my motives were otherwise.)

I have made no claim that is not supported by the Dawkins review of Behe's book or the article I quoted from the May 15, 2007 issue of PNAS. So, if you think I'm wrong, do as Dawkins suggests that Behe does and let peer review be the judge. If you think Cyborg is wrong or the myriad of other people on this forum who have reiterated what I've said in various ways or with their own better flair, don't insult them--save your energy to fight the whole world of scientists who are saying the same thing.

No peer reviewed scientist is saying that "natural selection is a random process", a "stochastic process", nor a "probable process", nor are they claiming that "evolution IS random".

On the other hand, some are definitely saying that natural selection is a non-random process or that the process is "not random" in peer reviewed current science articles. Until you understand this and understand why --you do not really understand natural selection. That's what Talk Origins asserts. And I concur.

pwnage

(and mijo...you might want to look up the word ironic; you seem not to understand what it means--which is ironic itself, but your use of it was incorrect).

 

[sphenisc]

The process of evolution is clearly described as probabilistic here...


Evolution

 

[steenkh]

The process of evolution is clearly described as probabilistic here...

Evolution

Very funny

 

[mijopaalmc]

Uh....articulett:

Evolution is random for the very reason that "[h]aving a good hand, doesn't guarantee success, [a]nd some can win with a bad hand". The probability of winning is the number of times you win when you play a given hand with a given strategy out of the number of times you play given hand with a given strategy.

 

[jimbob]

But As far as I follow, articulett, and you can't descrinbe any traits as making an organism "fitter", because only the fitter organisms reproduce and the definiton of "fit" means that an organismer reproduces.

If you were dumped in an ocean you might get that point when you start to realise that your particular genetics don't help you too much there.

Yes, the probability of survival to reproduction is affected by the environment.

What is your point?

As I see it, Cyborg's POV boils down to 'the environment is incredibly complex, but selection is deterministic, and the "the fit" are selected, so where an egg is laid in a nest relative to its siblings can also determine its "fitness".'

Uh no.

Could you please tell me where I have misrepresented you?

From my POV, one can talk about fitnes, and indeed have a rigerous mathematical definiton of it. A trait makes an organism fitter if it increases the probability of that organism reproducing.

Real rigorous there homeboy.

homeboy

?

What is not rigorous about it? One can perform experiments, and assess these probabilities. And one can preform statistical analyses on traits in the field, and their correlation with reproductive success.

So tell me - how useful are your lungs for your reproductive success if I drop you in the ocean again? Do you have a variable for the probability of you being drowned by madmen?

That would be a random event. My ancestors did not have any evolutionary pressure to adapt to such a situation, but it would be a random event.

One can then perform statistical tests on populations to see if this is actually the case for any individual trait.

Um no. One cannot perform statistical tests. One can perform statistical analysis. You really have got this all back-to-front.

And as part of the analysis, one can perform statistical tests. If I want to assess the probability that two populations are different using the Chi squared test.

But I can in principle use my definiton to perform statistical tests on observations of populations to test particular hypotheses about how particular traits (and thus mutations) affect fitness in different populations.

Yeah, and I can analyse the likelihood of winning a Poker hand but I still have to play the damn hand before I know if I won or not.

Do you not get that a statistical model does not make the actual playing of the game statistical?

But to analyse the probabilities of winning it is a good start.

Because probabilistic is a good description.

No it ****ing isn't.

Dice are probabilistic.

Lotto is probabilistic.

Coin flips are probabilistic.

Does a single one of these things even come close to being comparable to evolutionary processes?

It's incredibly piss-poor as a descriptor.

The Poisson distribution is also probabilistic.

Occurrence

The Poisson distribution arises in connection with Poisson processes. It applies to various phenomena of discrete nature (that is, those that may happen 0, 1, 2, 3, ... times during a given period of time or in a given area) whenever the probability of the phenomenon happening is constant in time or space. Examples of events that have been modelled as Poisson distributions include:

The number of roadkill (animals killed) found per unit length of road.

The number of pine trees per unit area of mixed forest.


The number of soldiers killed by horse-kicks each year in each corps in the Prussian cavalry. This example was made famous by a book of Ladislaus Josephovich Bortkiewicz (1868–1931).

The number of V2 rocket attacks per area in England, according to the fictionalized account in Thomas Pynchon's Gravity's Rainbow.


The number of viruses that can infect a cell in cell culture.

The number of hematopoietic stem cells in a sample of unfractionated bone marrow cells.

The number of soldiers killed by horse kicks directly affected the survival odf those soldiers. Some would have failed to breed because of an early death due to this, which had been modelled as a probability distribution

No I am saying that traits improve or reduce the chance of an organism reproducing.

**** - and having a better Poker hand doesn't increase my chances of winning?

Yes, it increases your chances, but doesn't ensure victory.

That is my point.

Of the 4-million cod from a single spawning, the vast majority will not reproduce, those which do will be down to essentially random events. Traits that alter the probability of survival are still selected for, but I would argue that there is precious little difference between the fitness of most os the spawn, some of which reproduce and some of which don't. In this case the winning strategy is to try dealing 4-million packs of cards, and hope two win.

And it ****ing works.

So what's your point again? You don't like the fact that the game has a lot of inequity built in?

No, my point is that those organisms which reproduce are the ones that are lucky.

At spawning is there any important difference between a codling that will reproduce, and one that won't? In general there isn't one is just luckier. The odds of an individual codling reproducing must be about (1/1000000)

The reason that cod produces so many ofspring is because any that didn't became extinct.

The "unfit" are efficiently culled. So are most, but not all of the "fit".

If a trait increased a codlings chances of reproducing hundredfold, it would still only have a (1/10000) chance of reproducing.

If all the codling from a particular mating, then in the first generation one might get 200 reproducing and not two.

The numbers are big to compensate for the atrocious odds. Not design, not haphazard, but probabilistic.

 

[jimbob]

From what I've seen, yes, it is. I haven't reviewed your posts categorically, but I have seen them, and read most of them, and that is "what I've seen."

I will agree that evolution is not haphazard, that evolution is not disorderly, that evolution is not disorganized. But I will not agree that it is not random; it clearly is.

Thanks, I agree with that statment too.

Could it be that the differences we see here are a result of different perspectives? Meadmaker suggested back in post 1418 that I viewed evolution as non-random because I look at the 'unfit' genes, and I readily conceded that the selection of 'fit' genes are completely random.

I do not have the intellect to follow the mathematical explanations like the landscape presented earlier, but am I right that they describe the selection of 'fit' genes?

Do the parties here agree that the unselection of 'unfit' genes is non-random?

So highly probable as to be almost inevitible. So yes.

 

[cyborg]

Yes, the probability of survival to reproduction is affected by the environment.

What is your point?

**** knows. Whatever it is certainly ain't getting through to you.

What is not rigorous about it? One can perform experiments, and assess these probabilities. And one can preform statistical analyses on traits in the field, and their correlation with reproductive success.

And one seems to think such models mean dick as far as what is actually going on.

That would be a random event. My ancestors did not have any evolutionary pressure to adapt to such a situation, but it would be a random event.

People drowning people isn't a 'random' (god damn I'm getting tired of sloppy use of that word) event - it happens with some regularity.

Where's your damn variable in your rigorous model jimbob?

Yes, it increases your chances, but doesn't ensure victory.

That is my point.

Wrong jimbob - that's MY point.

YOU NEED PLAYERS AS WELL AS CARDS TO PLAY POKER!

As long as you spend all your time worrying about the damn cards you're missing the point that once you put some players behind the cards natural selection is deterministic. Last time I looked the world doesn't consist of probability distributions throwing dice to decide how they modify their variables.

Do you or do you not get the difference between what you are modeling and what is actually out there?

No, my point is that those organisms which reproduce are the ones that are lucky.

Did they roll some dice to determine if they won jimbob? Did they?

At spawning is there any important difference between a codling that will reproduce, and one that won't? In general there isn't one is just luckier.

That is why one must PLAY THE DAMN GAME.

There is no other way of measuring fitness. You want to be able to neatly place a gene into a category or 'fit' or 'unfit' and you cannot ****ing do that because a gene is not 'fit' or 'unfit' - its vector is. And its vector may be somewhere where all its lovely genetic information means dick. Just like if I have a set of pocket rockets and the flop comes down for a straight flush - doesn't matter if I have the best hand pre-flop, I don't any more.

The numbers are big to compensate for the atrocious odds. Not design, not haphazard, but probabilistic.

In your model. In the real world real and actual spawn have to be culled. That's deterministic. The chess game is played and position means something.

 

[jimbob]

Oh-- Let me second what cyborg said.

Jim-Bob--all natural phenomena...including thorns and meteors are "expected" to be honing the pool of potential replicators. They are an expected part of the selection process--that is why it's called Natural Selection. If DNA can't get into a successful vector in a successful environment,--it just can't make it to the next level. If the DNA is "lucky" there's lots and lots of it copied into lots of vectors in lots of environments...and maybe one of them WILL get replicated.

That is why I say probabilistic to try avoiding confusion with "haphazard" selection.

Has the environment altered randomly since life evolved on Earth? I would say it has.

If not "randomly", has it altered in a fashion that would have been unpredictible?

If you don't like talking about the past, how about, "Is it possible to predict the environment in two million years time?"

Organisms can only evlove to fit the environment(s) which their ancestors evolved in.

Organisms that have the most spawn tend to have the most grandspawn-- With or without meteors, tornadoes, tsunamis, etc. If it's part of nature it's part of NATURAL selection. Do you know, some living things survived meteors...some might have even been in meteors...and we've recently discovered stuff way, way underground and in ice and living in the bottom of the ocean feeding off the radiation from earth's core--and things living in boiling lava. These things evolved because they had what it took to get their info. copied in the environment they happened to find themselves in. Extremophiles. Look it up.

But the KT impact, which was unpredictible from looking at the previous environment, significantly altered the environment and caused mass extinctions.

Evolved via natural selection--not at all randomly. They may not be fit by your standards, but you sure as hell ain't fit by theirs. This is what we mean by saying the "randomness" you speak of is just noise...fuzzing up your model and understanding of a simple concept. It's PART of "natural selection" It's redundant to count it in your definition of natural selection. And it misses the more important part of the equation. Why you insist on this obfuscation is beyond me.

Extremophiles have evolved to fit their environment, and they are fascinating in showing how effective evolution is.

My favourite is Deinococcus radiodurans

Is it really such a complicated idea that natural selection is probabilistic, where the unfit are highly likely to get "culled", but those with slightly better odds (which is due to heritible traits) will reproduce preferentially?

Why I persist in this is because; when talking to someone without knowledge, but intelligent, it is simple.

And deals with the question "So was that particular cod which just spawned the most fit of its brood?"

My answer is, "possibly; it certainly wasn't "unfit", but some others might have just been unlucky. "

The probabilities can be very skewed. But they are still probabilities. And the environment changes randomly.

BTW:

and things living in boiling lava.

No, boiling water in volcanic springs, not boiling lava. Even the BSE prion protein breaks down at about 800°C.

 

[jimbob]

Cyborg,

Do you really think that aircraft evolved?

 

[cyborg]

Nope. Aircraft do not evolve. They sit around or they get flown around.

Aircraft designs are a different matter.

 

[mijopaalmc]

The long and short of articulett's posts on Ayala:

"Until you accept my definition or "random", which scientists use only in non-technical, narrative writing in peer-reviewed journals and sound very confused doing so as they cannot avoid discussing the probability and chance of natural selection in the same piece, I'm going to impugn against your intelligence and character. I am also going to ignore that you provided a rigorous mathematical definition of probability so I can continue to call algebra, your argument, or anything that I so choose "random" so that I can declare the definition your are useless and use the definition I deem appropriate because it fits my preconceived notion of what should be."

You really need to look at how the scientists who describe evolution as "non-random" use "probability" and "chance" in reference to evolution, articulett. The Ayala piece you are so eagerly cite uses "probability" four times, once in the same paragraph.

This is how natural selection works: Individuals that have beneficial variations, that is, variations that improve their probability of survival and reproduction, leave more descendants than individuals of the same species that have less beneficial variations. The beneficial variations will consequently increase in frequency over the generations; less beneficial or harmful variations will be eliminated from the species. Eventually, all individuals of the species will have the beneficial features; new features will arise over eons of time.

The resistance of disease-causing bacteria and parasites to antibiotics and other drugs is a consequence of the same process. When an individual receives an antibiotic that specifically kills the bacteria causing a disease—say, tuberculosis—the immense majority of the bacteria die, but one in several million may have a mutation that provides resistance to the antibiotic. These resistant bacteria survive, multiply, and spread from individual to individual. Eventually, the antibiotic no longer cures the disease in most or all people because the bacteria are resistant. This is why modern medicine treats bacterial diseases with cocktails of antibiotics. If the incidence of a mutation conferring resistance to a given antibiotic is one in a million, the probability of one bacterium carrying three mutations, each conferring resistance to one of three antibiotics, is one in a quintillion (one in a million million million). Even at the peak of infection, when billions or trillions of bacteria exist in a sick person, it is not likely, if not altogether impossible, that any bacteria resistant to all three antibiotics will occur in any infected individual.

Natural selection is much more than a ‘‘purifying’’ process, for it is able to generate novelty by increasing the probability of otherwise extremely improbable genetic combinations. Natural selection in combination with mutation becomes, in this respect, a creative process. Moreover, it is a process that has been occurring for many millions of years in many different evolutionary lineages and a multitude of species, each consisting of a large number of individuals. Evolution by mutation and natural selection has produced the enormous diversity of the living world with its wondrous adaptations.

Natural selection accounts for the ‘‘design’’ of organisms because adaptive variations tend to increase the probability of survival and reproduction of their carriers at the expense of maladaptive, or less adaptive, variations. The arguments of intelligent design proponents that state the incredible improbability of chance events, such as mutation, to account for the adaptations of organisms are irrelevant because evolution is not governed by random mutations. Rather, there is a natural process (namely, natural selection) that is not random but oriented and able to generate order or ‘‘create.’’ The traits that organisms acquire in their evolutionary histories are not fortuitous but rather determined by their functional utility to the organisms, designed, as it were, to serve their life needs.

Basically despite using the rich language of probability to describe evolution and natural selection, Ayala declares natural selection "not random".

Is it any wonder that some people are confused by how biologists talk about evolution by natural selection?

 

[cyborg]

You still not get the difference between analysing the statistics of Poker and playing a game do you?

 

[mijopaalmc]

You still not get the difference between analysing the statistics of Poker and playing a game do you?

Could you explain the difference?

 

[cyborg]

Once the deck is shuffled, the players have anted up and the cards are dealt the role of randomness ends.

What cards will come on the flop is determined.

What cards the players have is determined.

How the players decide to play their hands is determined by the players.

The game is in play and all statistics will allow you to do at this point is have an idea of how it will unfold. How the game actually unfolds is deterministic.

 

[articulett]

The process of evolution is clearly described as probabilistic here...


Evolution

For the record, I don't believe in the Triassic either, but I do believe in the onion.

 

[articulett]

Is it any wonder that some people are confused by how biologists talk about evolution by natural selection?

Yeah, creationists, Jim-bob, and Schneibster.

What was your goal of starting this thread again?

And Jim Bob, life forms evolved on planets where "random" things like meteors occur, lightening, tornadoes, etc. Some even have adaptations where such events give them a survival or reproductive advantage... Things in the environment aren't random in the way mutation is random. Everything in the environment plays a role in determining what survives to reproduce--everything! Life on this planet evolved to live on this planet. If it can't keep going when something new event comes along, it doesn't.

You are confusing a statistical model of the game with the game itself and then congratulating yourself for being clear when no-one thinks you are being clear as far as I can tell. I don't think you, Mijo, or Schneibster understand each other or even have a clue why selection is non-random in comparison to the relative randomness of mutation. You change your definitions as you go.

I think everyone understands this much. You 3 think it is wrong to call natural selection "nonrandom" or "not random" and you think that you are using random in some singular correctly defined way. None of you have provided evidence that this is the case. You guys provide evidence where probability is discussed and pretend it is saying that natural selection IS random or that evolution IS random despite repeatedly being informed that that such conclusions are uninformative at best, misleading at worst, and identical to creationist obfuscation.

If you think you really are correct and that Dawkins, Ayala et. al. are wrong, see if you can get PNAS to publish your letter. At least get one respectable source that says "natural selection IS random" before you expect intelligent people to conclude that saying so is useful on any level. Because to the majority, it's convoluted gibberish on par with Behe-speak. Or if you think the Ayala paper defined random wrong--go tell them, not me. Tell talk origins and Berkeley too because they define the term as well.

Random components do not a random process make. Randomized double blind studies are not random processes themselves. They contain random components, but it would be misleading to call such trials "random", because they are a tool for getting understanding and order from the randomness--a way of subtracting the "noise". Natural selection is similar. Something which sorts the randomness is not itself, random! (Unless of course you are a creationist, obtuse, or desire to obfuscate.) Mutations have many things which influence it, but it is generally unpredictable. Selection is entirely determined by the replication success of the information (including mutations) in whatever vector and environment they find themselves in. If it's part of nature, it's part of natural selection. You have to actually play the game before you get the results. The randomness you speak of is included in the definition just like "double blind studies". Really. It's the same. Double blind studies aren't "random" though they contain randomization to bring about order...but of course, I realize that anyone who has not absorbed that by this point can not probably ever absorb that simple point. They either have an egotistical need to be "right" (at the expense of being clear, I guess) or a need for evolution to be random.

(Still no peer reviewed current scientific paper that says "natural selection is random", eh, Mijo?)