What evidence is there for evolution being non-random?

The creationists are mislead because they do not understand the part between randomness and complexity.... selective pressures and time. Artificial selection is a good way to start since we don't see "wild cows"... we can say why...and how they were created...the same goes for dogs...even though we didn't see it happen...we know for sure that they come from wolves because they can still produce viable offspring from wolves... Once you can get a picture of how such traits aided survival--humans fed and cared for the animals with the traits we liked best (and cultivated the crops we found most favorable)--and then honed and refined the best of this best (via breeding/agriculture) through eons. We now have a cow--a big docile good tasting tame food item that eats very cheaply and makes a lot of meat. Nature wouldn't select something so docile, of course...there would be some fat carnivores if cows roamed the plains...Leopards and hyenas could afford to get slower...and fatter...kind of like...well...humans.

Instead, the prey develop defenses and the predator evolve strategies to outwit those defenses... in an arms race of sorts--or they perish. Thinking of selection as random isn't really helpful. Think of artificial selection and extrapolate from there. I wouldn't use random in regards to explaining anything about evolution except in regards to mutation. But even then--there are trillions of mutations...only one has to be "selected" to move evolution "forward". Just one. It's not an efficient process. It can be a cruel process. But it is the reason for all the life we see around us. Understanding this allows us to ask the right questions as to "what it was" that made a particular trait an evolutionary asset. Sexual selection has some interesting "side effects"--and we can understand why-- e.g. why the peacock has such an elaborate tail. It wasn't random...it was peahen selection through time.

I think there is something so cool at being able to understand this and not have to make up explanations. I want everyone to understand it. We are the first humans in history who have such an opportunity and in a way so much more detailed than people past thanks to genomics. I don't know how to break the "top down design" mindset; I prefer to address it before it becomes entrenched, so "bottom up" thinking regarding complexity is natural. Such understanding is the key to understanding a lot of mysteries. In the future, the internet will be such a mystery to be pieced together...it evolved from the time humans hooked the first two computers together. We can trace that evolution...usenet...netscape...aol...yahoo...google...youtube...streaming...podcast...forums...etc.
Please communicate this information as clearly as possible.

oh--and Jimbob--having equal probability does not lead to uniform distribution-- If you flip a coin 6 times you have 64 combinations of coins that are equally likely. Only two of those are uniformly distributed--hththt and ththth (h=heads; t=tails). The rest are not. The uniformly distributed is the exception rather than the rule. It is how we tease out codes, meaning, algorithms, chaos, and patterns from randomness. Uniformity of distribution is suggestive of non randomness. If you randomly sprinkle pepper on your eggs, and it came out evenly distributed where all specks were equidistant from each other, I suspect you'd have achieved something on par with winning the MDC :) --(moreover, I don't think it would convey useful information to refer to seasoning food as a "random process", do you?)
 
Last edited:
articulett said:
oh--and Jimbob--having equal probability does not lead to uniform distribution-- If you flip a coin 6 times you have 64 combinations of coins that are equally likely. Only two of those are uniformly distributed--hththt and ththth (h=heads; t=tails). The rest are not. The uniformly distributed is the exception rather than the rule. It is how we tease out codes, meaning, algorithms, chaos, and patterns from randomness. Uniformity of distribution is suggestive of non randomness. If you randomly sprinkle pepper on your eggs, and it came out evenly distributed where all specks were equidistant from each other, I suspect you'd have achieved something on par with winning the MDC :) --(moreover, I don't think it would convey useful information to refer to seasoning food as a "random process", do you?)
Click to expand...

Unfortunately, this is not what mathematicians and statistician mean when they say a random variable is uniformly distributed. A uniformly distributed random variable is one that where a possible values of the variable are equally likely to occur. In other words, for a collection of n discrete events (e.g., coin tosses or die rolls) each event has a probability of 1/n of happening. This is what creationists mean in part when they say that evolution is random and this is why, according to creationists, evolution cannot occur. Uniformly distributed probabilities do not allow for one individual to be selected over another, which is what allows the frequencies of the favorable genes to increase in successive generations. Non-uniformly distributed probabilities are therefore the essence of natural selection as they favor fitter individuals producing greater numbers of offspring and less fit individuals producing smaller numbers of offspring, thereby increasing the frequencies of favorable genes and decreasing the frequencies of unfavorable genes in successive generations.
 
articulett said:
This is a common creationist claim which really shows a cluelessness about selection via time... the builder of complexity.
Click to expand...

Paul (or someone else knowledgable in biology),

Can you tell me if this characterization, central to articulett's discussion, is the prevailing explanation when biologists explain evolution and the role of natural selection? What I mean is the characterization of natural selection as the builder of complexity.

I want to describe what I think articulett's argument is. I want to do this because I want to make sure that I'm not accidentally constructing the infamous "straw man". Articulett seems to be saying that random events alone could not be responsible for the complexity of life. It is required that natural selection, a non-random process, be employed in order to build the complexity. (Note: I'm not going to try to be very specific about the definition of "random" in that description. I think the context makes clear in what sense the words are being employed.) I think I got that right, and that I'm not misrepresenting her argument.

I also think I'm right in saying that her argument would be extremely common among biologists. It sounds very much like Dawkins' brief description in "The God Delusion". It has been a while since I read a book specifically on evolution, but it sounds like what I've read in the past as well.

The reason I'm asking is because the argument, at least as I have understood it, is wrong. And it's not just a little wrong, it's completely and totally wrong. It's not even all that difficult to see why it's wrong, although I confess I didn't notice that it was wrong until I was eating dinner last night. (I'm on a business trip, so was eating alone and my mind wandered to this discussion and to evolution.)

The problem is that along comes a mutation. Great. For the sake of argument, we'll assume that it's a mutation that increases the complexity of the mutated organism. Now, that organism might survive to pass on its genes to the next generation, or it might not. If it happens to make it more fit for the environment it finds itself in, that gene just got more likely to be seen in future generations. If not, the carbon atoms are about to be recycled into an organism whose genes got luckier, and we won't see copies of that gene in the future.

The thing is that natural selection didn't buld the complexity. In fact, it destroys some complexity. It never adds. It only subtracts. It's the mutations, the random parts, that create the complexity. Natural selection just prevents the wasting of carbon atoms on on forms that might be complex, but whose complexity doesn't fit in with the environment. It's the mutations that create the complexity, some of which are good, and some of which are bad for the future survival of the mutant.

Imagine if there were no selection. Imagine if everything survived. If that were the case, then life would be much more complex, because complex organisms that didn't fit would still survive and pass on their complex genes, which would mutate further into still more complex forms. The only problem is that you would rapidly run out of carbon atoms. Natural selection focuses the process toward those forms which happen to match the environment, but it cannot actually generate the complexity.

(In general, the process of making an organism match the environment will tend to produce greater complexity with time. If anyone doesn't understand why, I can explain.)

I suspect that if I were to go back and read "The Blind Watchmaker", Dawkins would clear up what he meant and it wouldn't be the case that he really believes something that is just plain wrong. In fact, that restaurant was in a shopping mall, and on the way out I stopped at the bookstore and glanced at a copy, and in his discussion of "biomorphland" he seemed to be saying much the same thing. However, it's interesting to me that his simplified discussion in "The God Delusion" would lead to such a commonly held, and wrong, interpretation.
 
Last edited:
Meadmaker said:
Articulett seems to be saying that random events alone could not be responsible for the complexity of life.
Click to expand...

I think it would be more correct to say that mutation, recombination, etc. are not sufficient to explain everything that we see in the actual world.

Apropos of Kleinman's arguments, selection reduces variability. So, in a sense, selection slows "evolution" -- provided one defines "evolution" as a reference to the emergence of new forms. That is not really the definition of evolution that most biologists use, however, which is one of many reasons why his arguments are misleading.

Selection is one of the prime reasons why there is order within the biological world. A world with no selection, in which every organisms that could survive does survive, would be a blooming buzzing mess. If infinite resources were available then every sort of possible lifeform would eventually surface and, by definition, survive. What selection does is limit this blooming mess to a relative few, which provides more orer to the entire system. Mutation, recombination, etc. is the creative force. Selection is the limiting force that 'creates' the order within the system of life. As a result, what we see in life does have a sense of order to it -- order provided by selection.

The other constraint on the system is the biological means of information gain and storage -- DNA and protein expression. Those two aspects -- selection and the structure of DNA -- limit the expression of possible lifeforms, so that we do not see pure chaos.
 
Just my 2 cents on an issue that has just surfaced...

Many -if not most- people tend to see humans as "more evolved" (and "more complex") than the rest of the living beings. Mammals would be more evolved than reptiles, reptiles more evolved than amphibians, and so on. Its an old view "the pinnacle of evolution", that has some similarities with the "apex of creation". There would be, in their minds, a drive towards complexity. Evolution would be equal to increase in compexity, and therefore non-random. It would have a "goal", to produce complex sentient beings. This reasoning is quite often used by theists to inferr some sort of design, following the anthropic principle lines.

This is just wrong. Evolution is adaptation. Nothing more nothing less. The way our planet's history unfolded so far, the interplay between biological organisms, geology and astronomical phenomena so far favored diversity and some complexity. Diversity of ecological niches, diversity of habitats, diversity of species and the development of a number of complex multicellular species.

For most of our planet's history, humble unicelular and simple multicellular beings ruled. And they still rule! In doubt? Check their biomass and compare with vertebrate's biomass.

But wait, there's more! In the future, as soon as our Sun starts to bloat, this diversity will start to fade away. Ecosystems and niches will crumble, but evolution, as long as there is a single living being in this planet, will continue. The environment, however, will favor the humble, unicellular or simple multicellular beings. So much for evolution=complexity. Evolution=adaptabillity. And if the beings with the better odds to survive are simple ones, they will be the selected ones.

Complexity or simplicity, as an evolutionary outcome, is only a function of efficiency to face environmental pressures. The more efficient (complex or simple) individual or species will have an upper hand (or tentacle, pseudopod, hoof, claw, wing, whatever). Note that when we think about how resistent and adaptable microorganisms are, complexity does not seems to have the upper hand. Complexity seems good for some specialized niches, but bad as a strategy focused on the survivabillity of a species in a varied number of environments and through varied enviromental changes/pressures.
 
Meadmaker said:
...The thing is that natural selection didn't buld the complexity. In fact, it destroys some complexity. It never adds. It only subtracts. It's the mutations, the random parts, that create the complexity.
Click to expand...

Perhaps it would be more accurate to say "variability" rather than "complexity"?
 
articulett said:
oh--and Jimbob--having equal probability does not lead to uniform distribution-- If you flip a coin 6 times you have 64 combinations of coins that are equally likely.
Click to expand...

That is a binomial distribution. And is not as you describe:

Probability of all heads=(0.5)^64
Probability of all tails=(0.5)^64
Probability of 63 heads and 1 tail={64*(0.5^63)} if I haven't made a stupid mistake... There are 64 "locations" for the individual tail coin.

Picking cards once from a pack, or rolling a single n-sided die once will give a uniform distribution.

Only two of those are uniformly distributed--hththt and ththth (h=heads; t=tails). The rest are not. The uniformly distributed is the exception rather than the rule. It is how we tease out codes, meaning, algorithms, chaos, and patterns from randomness. Uniformity of distribution is suggestive of non randomness. If you randomly sprinkle pepper on your eggs, and it came out evenly distributed where all specks were equidistant from each other, I suspect you'd have achieved something on par with winning the MDC :) --(moreover, I don't think it would convey useful information to refer to seasoning food as a "random process", do you?)
Click to expand...

That is to do with entropy. And the joys of thermodynamics. The structure you describe is ordered, indeed "crystaline" if you mean "equidistant from their nearest neighbours" Otherwise all the pepper grains would be in the same location.

I would guess that one could probably use a gaussian distribution to model the distribution of pepper grains, and I can conceive of times when this would be useful:

An automatic seasoning machine in a production line where the x location of the seasoning is described by a gaussian with a mean and standard deviation, and also the y-location. One might want the spread to be wide enough to cover a sertain width and not waste too much. For example​

It is a percentage game. But I would still say random. If it was not, then the evolution of humanity would have been inevitable.

Maybe playing whist with an infinite pack, where the lowest cards in each hand are discarded, along with a random selection and number of the other remaining cards in the hand. A random number of these cards can then be replaced by a random selection of cards with a mean value equal to the remaining cards in the hand.

If someone can understand that there are different odds in card games or backgammon, then why not in evolution?

This would give a drift upwards in value...

and replaced randomly from a population with a mean and spread described by the remaining cards in the hand. And where also from the
 
jimbob, I believe the prob of getting 53 heads and one tails is 0.5*(0.5^53), or in other words, the same as getting all heads or tails.
 
Ichneumonwasp said:
Yes, correct. It depends on the type of selection.
Click to expand...

Heterozygote advantage ftw.

Actually, it's more then that, though. Neutralist theory is not enough to explain the high level of polymorphisms we observe in populations. Neither is heterozygote advantage on its own, as it is fairly rare at loci. Some form of selection must be happening, but it is a bit of a mystery as to how it works. :)
 
Taffer said:
jimbob, I believe the prob of getting 53 heads and one tails is 0.5*(0.5^53), or in other words, the same as getting all heads or tails.
Click to expand...


No, because you have 64 throws, and any 63 of those may be the heads. If you had 63 throws the odds would be (0.5^63). There are 64 places for the tail, so this has to be multiplied by 64.
 
jimbob said:
No, because you have 64 throws, and any 63 of those may be the heads. If you had 63 throws the odds would be (0.5^63). There are 64 places for the tail, so this has to be multiplied by 64.
Click to expand...
I think your probabilities are still off. The chances for any exact combination is the same any other. In other words, the chances for throwing exactly 63 heads and 1 tail is the same as the chance for throwing exactly 64 heads, or exactly 64 tails, or exactly 36 heads and 28 tails. Now, if you looking for the probability of throwing at least 1 tail, that does improve the chances, but looking for exactly 1 tail does not.
 
Vorticity said:
Perhaps it would be more accurate to say "variability" rather than "complexity"?
Click to expand...

"Complexity" is a difficult thing to define. The only rigorous definitions I have seen involve information theory. The more bits are required to define something, the more complex it is. In that sense, humans are indeed more complex than bacteria, or than lobsters.

So, I really did mean complexity. There is a second phenomenon which results in the diversity of life, because from a common ancestor, life will diverge in multiple directions, but that's a different thing.

The creationist argument is that the complexity of life couldn't have arisen by chance. The counterargument is that natural selection isn't chance. It's an ineffective counter, because natural selection doesn't create complexity. The accurate counterargument is that slow, gradual, change can produce complexity, even from random variation. Natural selection can then throw out poorly adapted forms, regardless of complexity. Natural selection can be responsible for increasing the average complexity of the population, but it can't be responsible for creating something complex.

Skimming through "The Blind Watchmaker", which I last read shortly after its initial publication, I was reminded of just how good of a writer Dawkins could be. I wish he would stick to evolution, and ditch the philosophy.
 
Last edited:
I less than three logic said:
I think your probabilities are still off. The chances for any exact combination is the same any other. In other words, the chances for throwing exactly 63 heads and 1 tail is the same as the chance for throwing exactly 64 heads, or exactly 64 tails, or exactly 36 heads and 28 tails. Now, if you looking for the probability of throwing at least 1 tail, that does improve the chances, but looking for exactly 1 tail does not.
Click to expand...

No, JimBob is right. Let's try it with 3, instead of 64.

HHH
HHT
HTH
HTT
THH
THT
TTH
TTT.

How many have 0 tails? That would be 1. How many have exactly 1 tail? That would be 3. How many have at least 1 tail? 7.

It's the binomial distribution, all right.
 
Meadmaker said:
No, JimBob is right. Let's try it with 3, instead of 64.

HHH
HHT
HTH
HTT
THH
THT
TTH
TTT.

How many have 0 tails? That would be 1. How many have exactly 1 tail? That would be 3. How many have at least 1 tail? 7.

It's the binomial distribution, all right.
Click to expand...
Ok, this is right. I was thinking of the probability of an exact pattern, HHT is exactly as likely as HHH, or TTT, or HTH, or TTH.

ETA - Which looking back to articulate's post, is exactly what she was referring to. With 3 coins there you have exactly 8 combinations, and each is exactly as likely as the next. The way you and jimbob are looking at it, the 3 coins would only yield 4 combinations: 3H, 2H 1T, 1H 2T, and 3T.
 
Last edited:
Davidjayjordan said:
All biological study shows that evolution is not random but comes from design. Study biology and note that impossibility of one evolving from another. get serious and don't believe the magicians of evolution. Study the specific systems in the bodies of different species and note their differences and instincts as a whole and as separate identities.

Evolution is impossible the more you study and learn.
Click to expand...
(emphasis mine)

I just thought I'd point this out to the people who the creationists only argue that evolution is random. Here is an argument that possesses the classical creationist trope that structures cannot evolve one into another but depends on the such "biological study" reveals that everything is designed and therefore is not random. Now, I know that Davidjayjordan generally incoherent and that he may only represent his own unique take on all things; after all, he seems to acknowledge that evolution actually exists. However, it does not seems that creationists' arguments necessarily rest on evolution being random. Creationists instead seem to argue whatever most benefits their stance at the time of the argument. Thus, I don't see why it is so necessary to avoid "random" in descriptions of evolution, at least for the sake of not confusing the creationists, since creationists (or at least one creationist) seem to argue evolution is not random to suit themselves.
 
mijopaalmc said:
(emphasis mine)

I just thought I'd point this out to the people who the creationists only argue that evolution is random. Here is an argument that possesses the classical creationist trope that structures cannot evolve one into another but depends on the such "biological study" reveals that everything is designed and therefore is not random. Now, I know that Davidjayjordan generally incoherent and that he may only represent his own unique take on all things; after all, he seems to acknowledge that evolution actually exists. However, it does not seems that creationists' arguments necessarily rest on evolution being random. Creationists instead seem to argue whatever most benefits their stance at the time of the argument. Thus, I don't see why it is so necessary to avoid "random" in descriptions of evolution, at least for the sake of not confusing the creationists, since creationists (or at least one creationist) seem to argue evolution is not random to suit themselves.
Click to expand...

This was already mentioned by Whitey-- and the whole idea of characterizing the scientific position as "complexity by chance" is to contrast it with design (top down). Life is "designed" from the bottom up. It's not preplanned, it's modified as it goes--sort of like the internet. No one conceived it in it's present point, no one knows where it's going, everyone who participates is part of the evolution of the internet. And all life forms are part of the evolution of life if they manage to get their genome copied. Sure, evolution "prunes"--but pruning brings order to "chaos" or "randomness" or the "miscellaneous." In this way, it is the opposite of "haphazard"--the creationist "straw man" characterization of evolution.

And I just don't speak engineer...I'm sorry. Life forms have a quality different than gases and stochastic processes... reproduction and death-- in this way life is different and these are the mechanism by which selection acts.

I don't care if you avoid random or use it or whatever. Biologists say "random mutation coupled with natural selection" which as the clearest summation of evolution. You don't have to worry about definitions and technicalities. Natural selection brings order from the bottom up--there is no need for top down "design". That is in important concept to get younger people to understand, so that the creationist canard mentioned on the first page doesn't readily mean "goddidit". For older people, I wouldn't even bother attempting to change their thinking. I'm not sure it can be done.
 
jimbob said:
No, because you have 64 throws, and any 63 of those may be the heads. If you had 63 throws the odds would be (0.5^63). There are 64 places for the tail, so this has to be multiplied by 64.
Click to expand...

If you mean a precise sequence of 63 heads and then one tails, the probability is identical to 64 heads.

However, if you mean a sequence of 63 heads and one tails, then you are quite correct and nevermind. :o
 
You must log in or register to reply here.

Back
Top Bottom