• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Randomness in Evolution: Valid and Invalid Usage

mijopaalmc said:
It is interesting that no-one who think evolution is non-random has answered lenny's questions.
Click to expand...

so talking to folks here in oxford (UK), it seems this reflects a tactical choice which, to me at least, seems strategically counterproductive. the central role of randomness is denied rather than explained, even though this role is fully accepted, rather wellunderstood, and poses no fundamental challenge to "natural selection".

i am told dawkin's discusses this often in lectures noting that it can be counterprodutive, but i have only heresay evidence this is the case. surely he himself must have written about it somewhere? no?

perhaps this tactical approach is useful in some places (?the US?) where naysayer confusionist arguements might make shortterm gain (as in "see: they say it is just random"), but in the long run surely this tactic will prove counterproductive. no?
 
lenny said:
i would really like to understand all the confusion/argument here.

articulett, statisticians would generally not be "smart people" by your definition. random processes (aka stochastic processes) are indeed considered random, you seem to think this is a very restrictive condition, for the most part they do not.

not all random processes are IID (independent, identically distributed) or Gaussian distributed, "most" are subject to "deterministic" influences; they are random processes merely because there is a random term (in discrete or continuous time) onthe right hand side of the equations that define them.


do we all agree that a classical random walk is a random process?

and that so is a random walk on a tilted surface (that is a random process that contains a deterministic drift?)

if you agree the second is random, i do not see why you reject the idea that evolution is random, (given the typical alternatives: random and deterministic, would you want to say evoluiton is deterministic?)

articulett, you just seem to allergic to the use of the word "random" in connection with evoluiton, would you consider it OK to call a radiation induced mutation random?

cyborg could you tell me which post you explained things in, there are over a thousand on this thread!

thanks.
Click to expand...
Hi lenny:
The confusion comes because certain posters are trying to apply their interpretation of the word random (which seems to be "non-deterministic") to the entire theory of evolution.

The "random walk on a tilted surface (that is a random process that contains a deterministic drift?)" bit of your posting is actually applicable to a simple description of evolution as a random walk on a fitness landscape.
The random walk part comes from the dozen or so different kinds of mutations that can happen in genes.
The "fitness landscape" is natural selection.

People tend to object the description of evolution as "random" since that implies that evolution is not predictable or repeatable. But evolution is predictable (e.g. given an environmental niche we can predict that there will be something to fill it; given a gap in the fossil record we can predict that there was something to fill it) and repeatable - see the E. coli long-term evolution experiment.

I would say that evolution is not random but it is also not deterministic. It is something in-between. You could call evolution "mostly deterministic" given the overwhelming effect of natural selection.
 
So this stultifyingly idiotic zombie discussion shambles on...

lenny said:
do we all agree that a classical random walk is a random process?
Click to expand...

It's a type of random process, yes.

and that so is a random walk on a tilted surface (that is a random process that contains a deterministic drift?)
Click to expand...

That's called a "directed random walk". It is not called "random". See the difference?
 
mijopaalmc said:
No, and I surprised that you don't see the equivocation (or the internal logical contradiction if you are not equivocating) in your statement.
Click to expand...

Back to this again... how wonderful.

So, mijo, tell me - are smoke detectors random? Yes or no?

Until you say one or the other with no caveats I'm not going to respond to anything else you post. :)
 
sol invictus said:
Back to this again... how wonderful.

So, mijo, tell me - are smoke detectors random? Yes or no?

Until you say one or the other with no caveats I'm not going to respond to anything else you post. :)
Click to expand...

As you have shown quite amply, one cannot answer that question without caveats when the person equivocates on the meaning of "random" one is using. The smoke detector is based on the random process of radioactive decay; however, since there are so many decay events the "no smoke" signal behaves in a predictable fashion, as dictated by the law of large numbers, a property intrinsic to the vast majority of random processes.
 
thanks Reality Check
Reality Check said:
The confusion comes because certain posters are trying to apply their interpretation of the word random (which seems to be "non-deterministic") to the entire theory of evolution.
Click to expand...

yes, i see this. i introduced the "random walk on a tilted surface" as a step toward stocahstic motion on a fitness landscape, naively assuming both sides would agree.

but i see why, for tactical reasons, one side goes quiet.

this may be an issue in public presentation of natural selection, but it is not an issue in statistics or in physics (or amongst researchers in natural selection speaking in private). from a pedagogical point of view it seems a high risk strategy not to simply explain that stochastic dynamics does not mean "utterly random".

but i have no desire to add fuel to the confusionist argument.

I would say that evolution is not random but it is also not deterministic.
Click to expand...

i understand. and i may understand why you would say that. nevertheless mathematically (and philosophically) a process is either determinsitic or not determinisitic (sets of measure zero excluded), and eventually i expect biologists will have to declare. i personally believe declaring early is better, but you guys are the ones fighting the fight. i hope understanding prevails in the end!

thanks for your reply.
 
lenny said:
this may be an issue in public presentation of natural selection, but it is not an issue in statistics or in physics (or amongst researchers in natural selection speaking in private).
Click to expand...

I'm not sure what issue you're speaking about, but if you think most statisticians and physicists would agree with the statement "evolution is random", I'm pretty sure you're wrong.

nevertheless mathematically (and philosophically) a process is either determinsitic or not determinisitic (sets of measure zero excluded), and eventually i expect biologists will have to declare.
Click to expand...

Really? Says who? To make that claim you will need to define "determinism". Please make your definition specific enough to answer the question of whether QM in the many worlds interpretation is deterministic. You should also give examples of at least one deterministic and one non-deterministic real-world process in order to demonstrate that your definition is useful in categorizing them.

Having done that, you will need to specify the definition of "random", which was the topic of this thread. Presumably that definition will explain why you have brought up the term determinism.
 
lenny said:
but i have no desire to add fuel to the confusionist argument.
Click to expand...

On the issue of strategy - that's largely a question of how best to communicate the concepts underlying evolution. I think the answer is to explain that there are random elements, like a large meteor hitting earth, and that there are non-random elements, like the effects on species of the climate resulting from that meteor strike.

In the case of those bacteria, which petri dish first evolved the ability to metabolize citrate was totally unpredictable, but the fact that that particular ability might evolve (out of an infinity of possibilities) was predictable.

I don't think anyone above chimpanzee level will have trouble understanding that a process can have both random and non-random elements. But when you say "evolution is random", you are communicating to most people the claim that there are no non-random elements, which is simply false.
 
I know it's probably futile for me to think that those who argue that evolution is non-random will think that anything I say clarifies my argument but I thought the following post clearly explains my thoughts about the "meaninglessness" of the definition of "random" that I use:

Vorticity said:
thaiboxerken said:
Essentially, everything is random according to your definition of what "random" is.
Click to expand...

In principle, yes.

And by the way, this is not "my" definition. This is the standard definition.

You've basically made the term "random" meaningless.
Click to expand...

Not at all. Just because everything is in principle "random" does not in any way mean that the term is without meaning. Many, many processes can be predicted using deterministic models with far, far more accuracy than we can possibly measure. It makes sense in these cases to speak of these processes as "deterministic", since it is beyond our ability to tell otherwise.

So, in practice, there is a kind of subjective "degree of randomness" that a process must have in order for it to be utilitarian to refer to it as "random", even if all processes are random according to the technical definition of the word. In other words, we can make a conceptual distinction between the two concepts which is very useful in practice, in much the same way that we can make a distinction between "circles" and "squares", even though no absolutely perfect circles or squares may exist in real life.

As for your drive to work, I think it does make sense to refer to it as random. Take car accidents, for example. There are good statistics available on the incidence of such accidents per car per trip. We can measure the probabilities involved, and they are not negligibly small. Assuming you have the same probability as anyone else of getting into an accident, then your successful arrival at work is indeed the result of random chance.

I can think of 2 or 3 times in the last year when I started driving to work and for whatever reason ended up not getting there that day. So - if we stipulate that the factors that kept me from getting there are random - then my probability of not getting to work each day is on the order of 1 in 100. To me, this is non-negligible.

And as for evolution, It seems clear to me that it is significantly more random than whether or not you get to work safely. Walter Wayne made a good point in the other thread:

Walter Wayne said:
A question for the "non-randomites". Do you think that the rise of humanity was inevitable, given the conditions when life first crawled onto land?
Click to expand...

Again, to re-re-re-iterate: Saying that "evolution is random" is in no way a criticism of evolutionary theory.
Click to expand...
 
Last edited:
Ha-- Mijo is shopping his "evolution is random" around other skeptic forums including IIDB and Richard Dawkins net... and getting his butt handed to him on a platter as he did here.

Mijo needs to "believe" that it somehow makes sense to call evolution "random"-- but no one is buying it.

Mutations are generally considered random in that they happen whether they benefit an organism or not-- SELECTION chooses from that randomess--it is the derandomizer and the best of the randomness is multiplied exponentially.

To quote Dawkins reveiw of Behe's book again:

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. http://www.nytimes.com/2007/07/01/books/review/Dawkins-t.html

Ha ha ha ha ha

No biologist is defiining random like Mijo is.... if you are going to use random in evolution, you better define as the peer reviewed scientists in the articles do if you want to convey understanding. In fact, in biology there are degrees of randomness. Of course this has been explained to Mijo in thread after thread for years and now on different forums because he's tired out the people here.

Mijo's goal is to prove to himself and others that evolution is random just like the creationist canard says -- which is exactly what as Behe is doing.

Kleinman, another creationist who needs to prove that "evolution is mathematically is impossible" is also shopping his same story across skeptic forums.

Myself, I'll stick with those who teach many and ARE successful at conveying understanding of evolution--not the self-appointed experts.
 
Just a quick question from the sideline, is there a scientific or mathematical name for the type of biased randomness we find in nature?
 
"natural selection"... Dawkins et. al. consider this the opposite of random-- the derandomizer. It DETERMINES what evolves and how. It is responsible for the appearance of design.

When a dice is biased, we call it a loaded dice.

Mijo's definition would call such a dice "random".

Not very explanatory.
 
The more I think about this, the more complex the question becomes. The history of life on the planet does seem to point to a bias towards greater complexity. In my mind the more complex an organism, the lesser the chance an individual mutation can make the particular animal unviable. Yet with a less complex creature, a single mutation could do the reverse.

So then is complexity a genetic insurance policy against genetic coding errors? If the answer becomes yes, then is correct to say that life is activily attempting to lessen acummulated randomness?

Or am I just over tired lol
 
Actually that's pretty good... life become more "efficient" like technology does in a way. However most of life on this planet is very very simple life... so that's a great strategy too-- but not the only way to get DNA passed on. Sexual recombination happens to be a super way to pass on huge amounts of DNA data into increasingly "complex" and efficient "DNA processors"... It ups the variety from which natural selection can act on... and just like the information in evolving technology is cumulative and more efficient (it doesn't go backwards) neither do genomes--the only way is "forward".
 
Last edited:
MG1962 said:
So then is complexity a genetic insurance policy against genetic coding errors? If the answer becomes yes, then is correct to say that life is activily attempting to lessen acummulated randomness?
Click to expand...

It's more complex and interesting than that. To a large extent organisms are able to control how many mutations they undergo (search p53 and cancer, for example). If there are too few mutations the species is at a disadvantage. For example a virus which cannot mutate rapidly to overcome immune defenses would quickly go extinct. On the other hand if there are too many mutations, not enough offspring survive and the species goes extinct. So there is an optimal rate.
 
Yes, parts of our genome and others are highly conserved (very resistant to change) and other parts are hot-spots for crossovers and mutations because because it's proven a good gamble evolutionarily.

A clever virus strategy is to evolve just enough to encourage your host to pass you on--but not so much as to kill your host. Colds are "clever" virus because they cause a reaction (coughing, sneezing, runny noses, etc. that make them very easy to pass on... and they don't harm their vector so that it can keep on passing colds in the future). But because the immune system recognizes old viruses and protects against them, they need to keep altering themselves a wee bit to sneak by and cause their vector to sneeze, cough, etc. again.
 
MG1962 said:
Just a quick question from the sideline, is there a scientific or mathematical name for the type of biased randomness we find in nature?
Click to expand...
mathematically, one usually just speaks of random processes (or equivalently stochastic processes), "pure randomness" being a special case.

almost all modelling involves what you called "biased randomness", since the "pure randomness" (for example, where a variable takes on random values which are independent and identically distributed (IID) draws from some given distribution.) is a very special case.

a standard text is:
The Theory of Stochastic Processes by D.R. Cox & H.D. Miller

and you can read the first page at:

http://www.amazon.com/gp/reader/041...RdH9yhvrAbt5YyOBQpNr3PA4nm8dBw0A=#reader-link

here the random walk variable is X, and since it has a memory it is arguably already a case of "biased randomness", but the steps in the random walk (the Z's on this page) are IID (that is "pure random"). as you might guess, most of the 407 other pages in this book focus on more interesting varieties of "biased randomness" - referred collectively as "stochastic processes".

so the biased/pure distinction fails to attract the same attention inside mathematical biology that it does in public discussions of evolution.

note that this is undergraduate mathematics, not rocket science:
From http://www.stats.ox.ac.uk/prospective_students/bammath_maths__and__statistics/course_details
In the third year, all Mathematics and Statistics students take the
compulsory Applied Statistics course, and at least one of Statistical
Inference and Stochastic Modelling
Click to expand...

to argue that the fact evolution is best modelled as a stochastic process (aka random process) implies "evolution is random" in the pure random sense is just silly.

i can see that it might be an effective tactic if someone aims only to confuse the issue. i am not sure what the best tactical response is.

(hope that helped at the sidelines)
 
You must log in or register to reply here.

Back
Top Bottom