My old pal Marco has begun a quixotic crusade against Ockham’s Razor, claiming that it has been used unfairly to
browbeat dissenters and prop up flimflammery. Exhibit #1 in his crusade is the
hypothesis that natural selection on random mutations is sufficient to explain
the staggering diversity of the natural world.
Okay, so there are such things as
horizontal gene transfer and the intriguing symbiotic relationship at the
beginning of eukaryotes. And trivially, some stretches of DNA will be less
stable than others, and some cellular environments will be more exposed to
mutagens than others, so mutations are trivially “non-random”. He insists this
is not what he means. There must be some “non-random” mechanism to generate “good”
mutations. I said, no, this is bogus. I said this at great length through a
long rambling argument in the comments on his blog before I got tired of saying
it, but having a pretty good stamina for argument I recovered after a few weeks and thought I would come back
and say it again here.
1.
Heritable random mutations happen all the time. The
environment of cellular replication is rife with things that can cause
mutations, the mechanisms intended to pick these up don’t always work, and if
you irradiate living things it is easy to generate non-viable mutant offspring,
because most change is bad. This is the
basis for every post-nuclear-holocaust movie ever and my political philosophy.
And yet:
2. Changes
that are good happen all the time. There are lots
of examples of this unfolding in real-time as we speak. Of course, these good
changes might not have anything to do with random mutations. But there is no *clear and pressing need* to postulate
any mechanism beside random mutation. If there is no clear and pressing need to
introduce a new parameter, we don’t do it, because then we are off chasing
will-o’-the-wisps all the time. This is why Ockham’s Razor is good. On the off
chance a random mutation makes an organism that isn’t non-viable, the random
change will propagate through the population.
3.
The 747 argument is bollocks. Marco is (I think)
convinced for a need for non-random mutations by the dodgy statistical
arguments of various creationists and steady-state-universists that random
changes are insufficient to explain dramatic changes in speciation, because there
isn’t enough time in the universe for little changes to add up to the big complicated
changes we see. This is assuming an oversimplified linear y = mx + b of what
living systems are like and how changing them works. Living systems are complex
systems where arbitrarily small changes on the molecular level can have
arbitrarily large effects on the macroscopic level. Think of all those genetic
diseases that can be traced back to one little residue on one protein being
skew-iff. The butterfly effect is the explanation for butterflies. And any
arbitrarily large change that is not too large to make an organism unable to
reproduce is allowed.
4. You
can only find out what the programme does by running the programme. There is no way for a system to know in advance that a change is
going to be good. You don’t know for sure when you add a new ingredient to your
omelet, no matter how many other foods you’ve added dried cranberries to and it
worked out fine. The Vice-Chancellor doesn’t know for sure when he brings in a midnight
to dawn teaching period, no matter how many highly-paid consultants recommend it. To find out what changing a line of code does, you need to run the
programme. To find out what changing the genotype does, you need to generate
the phenotype. There are no shortcuts.
So, there is no need (points 1-3) for “non-random”
mutations and there is no conceivable mechanism (point 4) for “non-random”
mutations.
This may seem like a series of disconnected
ex cathedra pronouncements. Very well, it is a series of disconnected ex
cathedra pronouncements. That is just for brevity. I am prepared to defend them
all to the death with copious citations and ninja logic.
65 comments:
Hi Chris, for some reason I've only just found this here. I do disagree at some level with most of the pronouncements. However my reasoning behind any rejection of any of the particular points, is connected with my rejection of other points. To top off the frustration, no scientist is really thinking along my lines so I could only have citations to people considered crackpots, which would hardly help my case. Your citations would be very unlikely to convince me either. I am internally comforted in the knowledge that I know I am right, and there is a reasonable chance I could be proved right within my lifetime.
I would change your point in 2.
But there is no *clear and pressing need* to postulate any mechanism at all that specifies whether mutations are random or non random. Evolutionary science would still give the same answers whether it was random, trivially non- random, or sophisticatedly non random via natural means.
random citation
Let me take these points one at a time.
1. I do not accept the premise that heritable random mutations happen all the time on two grounds. The ones that we can measure are on supposedly "junk" bits of DNA and just because we do not think there is a purpose in these and we cannot predict them, we assume they are random. Irradiating things does cause heritable and mostly damaging mutations, but again they are not provably random. And the fact that they are happening *because you are irradiating them* is more reason to think they are not random.
1(a) Since the variation in 'junk' DNA is consistent enough over time to give 'molecular clocks' without glaring inconsistencies, it appears to me that the rate of mutations in 'junk DNA' is essentially constant for organisms under very different evolutionary pressures, which is evidence to my mind that whatever is bringing it about is indistinguishable from randomness in any practical sense.
1(b)When you irradiate DNA, you break bonds and introduce defects. These are due to high energy particles generating free radicals which will react with DNA on the basis of local chemistry, which is many levels of organisation removed form function. There is no logic in your assertion that mutations are *not* random "because we are irradiating them".
2. Let it be clear here. The definition of "random" is quite important and it "may" mean to you that there is no overall forward thinking plan long term as synonymous with random. Or that anything not designed by an entity we can relate to or know about is "random". I think it is a statement of faith similar to "Evil people are going to hell" that "bad mutations will not propagate as effectively" Too much faith is placed on something as vague as "natural selection" to perform something so complex and information intensive as filtering good mutations from bad mutations, especially when many are crucial for rare events and may be dependent on other otherwise useless genes. In programmed situations mimicking evolutionary design the mutation, evaluation loop is controlled and is done repeatedly in software for the very advantages that you don't want to "run the whole program" to optimise one bit. Being random is not knowing which bit needs changing. It all points to sophistication of the type that is embedded in the software - ie. not random. These mutations are theoretically traceable back to their non-random source. This would be useful science.
Re: 1 (a)
My impression is that molecular clocks are functional molecular clocks. Junk DNA could very well be how it keeps track of which important geological, astronomical, or weather related catastrophic events happened when in the sequence of evolution. I find this a plausible alternative explanation for "junk dna"
I would agree with you entirely with you definition, if that is what it is, that 'Being random is not knowing which bit needs changing.' I would go to the wall for the statement that mutations in that sense are sufficient and there is no *need* to postulate any other mechanism.
'Bad mutations will not propagate as effectively' is not a statement of faith, it is how I would *define* a bad mutation.
The sorting process does not need to be complex and information intensive; that is the beauty of natural selection. You (the progenitor) run the programme a gazillion times: there are plentiful resources of energy for you to do so. Programmers do not want to put into the resources of modelling the entire life cycle of organisms and all the other ways they could somehow muddle through to reproduce, but this has nothing to do with how nature does it.
Re your comment on 1(a), what conceivable advantage to an individual would there in be in doing such a thing? I can sense our common axiomatic ground eroding to such an extent that we may no longer be able to coverse at all. :(
3. The 747 argument is stuck in the "creation vs evolution" argument dichotomy, and any similarity to my argument is incidental and should not be used to build a straw man. Just because Wright and Orville didn't have the 747 in mind when they were building their aeroplane does not make their design process require "randomness" nor does it require superhuman engineering skill to eventually get to a 747. Equally, with evolution just because I agree that evolution does not "look ahead" any more than perhaps the next species does not mean it requires randomness. Equally, it does not require divine intervention to get the variety of species we have.
The molecular clock is useful in an evolutionary sense to understand the nature and regularity of extreme events to avoid extinction. Thus instead of losing a gene that may protect against the cold because it hasn't got cold in a thousand generations, it keeps it in reserve in part of the population.
If you are saying that my speculation as to your motivations is incorrect, and you are convinced of a need for non-random mutations for some other reason, I welcome the clarification! If you could spell out in small simple words exactly why you think natural selection on random mutations is not sufficient to explain the evolution of life from a primordial protoplasmic globule, I would be most grateful.
I am not sure what you mean by the statement:
"You (the progenitor) run the programme a gazillion times"
In a naive understanding of natural selection you run the program each time you reproduce. For humans that is not many times each???
Sorry, by 'you' I was thinking of a coral organism or an insect or one of those sorts of fish that can lay 300 000 000 eggs a year. My apologies! For us humans, evolution is all cultural evolution now. :)
If you are saying that my speculation as to your motivations is incorrect
Remind me again what your speculation to my motivations is.
That was my reason for mentioning the 747 argument: Marco is (I think) convinced for a need for non-random mutations by the dodgy statistical arguments of various creationists and steady-state-universists that random changes are insufficient to explain dramatic changes in speciation, because there isn’t enough time in the universe for little changes to add up to the big complicated changes we see.
To answer the gazillion progeny question - Yes - the idea is to have progeny of that scale not just with different mutations, but also to have a geographical spread of progeny, and yes natural selection is occuring and is helpful in that kind of situation, and there isn't many better reasons to have so many. The point is, A random mutation is in the order 1 in 300 million possibility of being beneficial, there is absolutely no reason to allow your progeny to have random mutations bestowed on them if you can control the mutations in any way. Species that cannot control their mutations will die off in a truly Darwinian way.
Time is not really an issue. That it is possible at all is an issue. The paradigm of both abiogenesis and evolution is fixed to a set of unproven statements that may be impossible, but perhaps not provably so. If abiogenesis is impossible, special creation is impossible, assisted evolution is impossible and life as we know it existing at the birth of our universe also impossible, where does that leave us?
4. For completeness. As I have said, natural selection based on not just *different* mutations on a set of progeny, but also geographic distribution is an important tool in the survival of the genetic line. You can only find out which of your *non-random* mutated progeny will be in a better place with the most suitable genetic adjustments until you run the programme.
1(b)Irradiating a specimen that has dna and can reproduce is not the same as picking particular gene sequences at random and changing them. Basically it is not a controlled experiment and there is no way to track randomness with causality in any useful way to make your point. Ok. Brute force and large scale haphazard wholesale changes to DNA can happen in a high radiation environment, and mutations are often extreme and unviable. So what. Yes large scale "random" variation is bad. also almost random, or non-random large scale variation is probably bad. Not really sure if it helps or hinders your overall argument so what the hell, I can concede the point if you must.
I can sense our common axiomatic ground eroding to such an extent that we may no longer be able to coverse at all.
There really is only my axiom rejecting Ockham's Razor that I sense is causing all the problems. Your definitions of "random" and "bad mutations" make no sense to me.
If we want to continue this conversation, we should dwell on the definitions of these terms. I feel you are implicitly and subconsciously using Ockham's razor to expand ad hoc observations to generalise conclusions.
By 'random' I mean there is *no* link whatsoever between where in the genome a mutation occurs and where a change in the genome might occur that might be beneficial for it.
By 'bad mutation', I mean a mutation that will not propagate as effectively (i.e., be reproduced in the population) as well as the unmutated form of the organism.
I will reiterate: If you could spell out in small simple words exactly why you think natural selection on random mutations is not sufficient to explain the evolution of life from a primordial protoplasmic globule, I would be most grateful.
I really, really do not understand your motivation with all this.
You are too brief in your treatment of #4. My assertion is that some mysterious DNA godlet cannot make a 'non-random' mutation directed to produce a particular outcome, because it does not know what the outcome will be until the phenotype has been generated and has had a chance to interact with the environment.
Finally: Abiogenesis is possible. Special creation is possible. Assisted evolution is possible. Life as we know it existing at the birth of the universe is not strictly impossible, but would require the assumption of a malign and capricious God. We are talking about adaptation and speciation, and here the model of natural selection has proven itself vastly more *useful* than the alternatives.
Yes. We have to dig deeper still to see where we lose the common basis of argument, see.
My view on your definition of random is that the probability of a random mutation being beneficial for the long term benefit of the species is somewhere in the range of 1 in 300 billion. If there is any pre-processing of genetic information at the point of reproduction or some sort of "error correction" if you will which is more likely to pick up problem mutations over any other sort, your definition of randomness no longer applies, even if it is a trivial change in probability.
If there is any pre-processing of genetic information at the point of reproduction or some sort of "error correction" if you will which is more likely to pick up problem mutations over any other sort, your definition of randomness no longer applies
No, I don't think so. There are certainly error correction mechanisms which will correct problems in DNA replication, but this will not induce any link between where in the genome a mutation occurs and where a change in the genome might occur that might be beneficial for the organism.
My issue with your definition of a bad mutation is that it presumes that mutations that are damaging to the long term (say Millenia) will have a proportional and meaningful deficit generation by generation in the meantime, that would cause that bad gene to die out. Genes that are redundant, ie, that perform the same task as another gene would be ripe for the "sacking". These can take up extra resources such by your definition they would be bad mutations,in a limited resource environment redundancy would die out and the chances of getting it back are slim, and the vulnerability to extinction without that redundancy is much greater in the long run. I cannot reconcile this issue with your definition of bad mutation. It is in the same idea space to me as an idea that bad people die young, or that the death penalty stops the crime it is associated with. It doesn't gel with my idea of reality.
What I mean by Random.
Redundancy does die out, and organisms become more specialised, making themselves ripe for extinction. This happens all the time. Degeneration is much more common than complexification - lots of parasitic organisms lose functionality wholesale compared to their free-living ancestors. Forget my definition of 'bad mutation', if it is causing you trouble. I don't know that it is very useful. All I'm trying to say is that there is no way of knowing in advance what mutation will be 'good'. Your example of your issue with my definition is a perfect description of how evolution usually works.
No, I don't think so. There are certainly error correction mechanisms which will correct problems in DNA replication, but this will not induce any link between where in the genome a mutation occurs and where a change in the genome might occur that might be beneficial for the organism.
Let's dig deeper. Do we have proof of the exact process of "error correction" at the point just before reproduction? Your assumption that this process is just as likely in any part of the genome as any other to let an error through, is dubious to me. Certainly any error correction algorithm in the real world is better off if it makes sure critical pieces are as they should be, and that no effort or energy is wasted on non-functional parts. where the algorithm is told probably doesn't matter so much is where random mutations will occur the most. Calling them random at this point is devaluing the algorithm. The most interesting part about the mutations is the part which is NOT random in this "not quite random" scenario.
I'ts all in that book I sent you a few years ago. You did read it, right? :) The machinery of molecular biology is well established.
What I still *really* want to know is your motivation. Why don't you think natural selection on random motivations is sufficient?
*goes off to check Marco's blog*
As far as that book goes, there are a lot of error correction that is well understood, let's say most. In a statistical sense virtually all of them are very effective but not "perfect". The impression that I got was that the systems for somatic cells were different to those of germ line cells, which are not as well understood, especially in a higher animals sense. The germ line cells is, of course what we are interested in here. A lot of research that I've delved into dwells on male human reproductive issues. Infertility issues are hereditary, but the researchers appear to be at a loss to identify genes generally responsible. The book doesn't really delve into the statistical treatment of heritable mutations, which is what we are talking about in this argument. I have a related theory about sperm line cells - that the error testing code of the gametes renders the human infertile if there are too many bad mutations. Thus there is no particular genes that affect fertility per se, but a judgement by the DNA gods that this individual should not reproduce.
My motivation is mainly on principle. I know many generations from now, people will realise that Ockhams razor is a valid principle for engineering, but not scientific truth per se. It is not at all that the simplest explanation is usually right, but that the simplest explanation is easier to remember and make use of to make stuff. The standard evolutionary synthesis has Ockhams razor cuts all over it. Yes it is simple to remember that God and Magic have nothing to do with it structured like that, but there is really no advantage to science and creative thought of what is actually happening in evolution.
Your example of your issue with my definition is a perfect description of how evolution usually works.
Degeneration and extinction is explainable - continued viability is not statistically verifiable in the same way with random mutations as you perceive them. There is just not enough new mutations that are good compared to new mutations that will go down the path of extinction.
I think the pronouncement that all error correction (talking about germ line cells) happens at the molecular level and is independent of any bias that may affect which mutations pass on to the next generation is an ex cathedra pronouncement based on what we know about somatic cells. The error rate of somatic cells is higher and more measurable, but extrapolating to the mutations in germ line cells is not proof, and there is no reason why the statistical basis would even be similar.
It is not at all that the simplest explanation is usually right, but that the simplest explanation is easier to remember and make use of to make stuff.
Okay, I think I see now - and I might have years ago when we were talking about this before. You do not at all disagree with the title of this post, my statement that natural selection of random mutations is sufficient. You just reject the principle that because it is sufficient, we should assume it is right. Am I right?
I also reject the principle that we should assume it is right because it is sufficient; but I adhere to the principle of least action, that we should go on using it as a working hypothesis until such time as there is clear and pressing evidence that it is not sufficient.
Umm.. Not quite. I do not believe we should use it exclusively as a working hypothesis. It crowds out other ideas in toto. Despite your protestations to the contrary this working hypothesis defies the requirement of proof, yet requires any competing hypotheses, even ones with tiny differences to be proven beyond reasonable doubt.
I do also believe it to be impossible, which is bad for a working hypothesis. I don't have proof, but I believe the statistical evidence supports the assertion.
Since last time we discussed it in depth, I have hardened my heart against it, as I have seen how it's opened up a whole raft of alternatives for the origin of species, including living comet theory.
I do also believe it to be impossible
Ow, way to crush my shreds of remaining faith in dialogue. :( Why, for the love of God, why?
It all relates to a thought experiment(that led to my epiphany on comets), that I did a couple of years ago when I was trying to argue very strictly from my own Marconomic principles. The thought experiment involved every scientific statement or working hypothesis that used Ockhams razor in some way to come to a simplistic conclusion. In my thought experiment all these hypotheses are ruled impossible. Then where does it lead science in that case?
With regarding to randomness, whenever I simplify the situation, the statistics don't add up. The complex situation of reality just gives the non randomness somewhere to hide amongst unpredictable and unknown phenomena that, with a bit of sleight of hand is called random. The modern knowledge of molecular chemistry genetics does not help, because we are shoehorning new information into the same arguments and structure from Darwin's day. This rather than letting the evidence speak for itself. Thus, I am happy to go with the axiom that as it stands, the modern evolutionary synthesis asks us to believe the impossible. Anything that derives directly from the need for randomness is equally poisoned fruit, and should also be ignored.
In my thought experiment, abiogenesis is also impossible, as is "magic" or any direct will of God that does not happen through naturalistic chain of causality. Also that life as we know it existed at the start of the universe is also impossible. I request you to do the same thought experiment if you really want to understand where I'm going with this, or where it is coming from. Until you've walked in my shoes with these axioms even temporarily, we are unlikely to be able to discuss it much further.
1. Any mechanism that allows someone to come up with testable predictions is of course okay. Has your thought experiment led you any testable predictions? How would you test them?
2. Much, much more important than parsimony is the principle of uniformity. My bedrock principle is that any theory advanced to explain something within the universe in the historical sciences must be *not inconsistent* with what we know from the experimental sciences. You need to advance plausible mechanisms for how your theories could work that is consistent with what we know. You haven't done that, and until you do that, I will not take them seriously.
3. whenever I simplify the situation, the statistics don't add up What do you mean by this? Statements like this are why I brought up the 747 argument. What doesn't add up in the statistics?
4. I'm not going do a thought experiment of assuming abiogenesis is impossible, any more than I'm going to do a thought experiment of assuming Tuesday is a legume. Words to me have meanings, and the meanings they have preclude such an experiment. If this means we have to talk about something else instead, so be it :(
1. Yes they have. I think I suggested a few regarding my Lamarckism theories way back when. Already, the fact that stressed organisms mutate in an augmented way which I cannot reconcile with the word random as you describe it, is a lead-in point to potential experiments, albeit we have to stress organisms to perform them. Certainly that is the tip of the iceberg and only deals with the parsimony of the present post in question. Certainly, my predictions with comets are testable, and stretch was one which is a requirement of my living comet prediction, and stretch appears proven beyond reasonable doubt, although by all reckoning, contact binary and ablation are going down swinging, Geocentrism style.
2. Uniformity. For one, taking the assumption that abiogenesis is impossible (etc.) is in no way inconsistent with the experimental sciences. Advancing plausible mechanisms appears a non-sequitur to the uniformity principle. There are myriad mechanisms that are consistent with all we know and could be true if our fruits of parsimony are ruled impossible. The problem is fitting it into what we *think* science has proven or demonstrated that is actually fruits of parsimony. It is too much of a bar to demonstrate what I mean for you when you have the concepts of causality and generalisation so conflated just in our argument of the definition of randomness. I don't see the point of picking one of the myriad of examples knowing that you would reject it on a ground that I believe is the fruit of parsimony, and that you would say is established science.
3. My reckoning is that if the convoluted complex situation is to be reckoned feasible with random mutations, than a simplified situation that we can actually calculate the probabilities will demonstrate this. I feel that in a simplified, but realistic situation for the simplest of organisms, a calculation of 1 in 300 billion of a random mutation anywhere in the genome being a beneficial mutation for the long term survivability is a conservative figure. My reckoning is that if a simple situation cannot be explained statistically, than something way different than random is happening in the convoluted complex situation. I think the impossibility of random mutations being sufficient (with a myriad of feasible alternatives) is a reasonable axiom to make for a thought experiment at the very least.
4. I must admit that the assumption that abiogenesis must have happened was the hardest one to let go as part of my epiphany regarding comets. For several months, I was convinced abiogenesis happened in comets. Still, the thing that rang the least true was partly your chemistry arguments that pointed to the fact that without external influence (or higher metabolism), there would be no reason why it would happen. Abiogenesis just seemed to hard to be possible, and when I took the step to tell myself it was impossible, other possibilities entirely consistent with *experimental* science, thus uniformity just materialised in my head. I would be happy to run an argument that Tuesday is a legume for as long as you like. However, I perceive it like Winston's replacement of the word God with fairies. I believe this "axiom" of mine, and I believe it can be entirely consistent with uniformity. Pick Tuesday or any other axiom you like and I will play ball.
I think I need to find a better hobby than arguing with people on the internet, I am growing too old and cranky for this.
There are myriad mechanisms that are consistent with all we know
Advance one. Just one. Argue it logically. Demonstrate how a system could possibly 'know' what mutation would be beneficial without running the program.
It is too much of a bar to demonstrate what I mean for you when you have the concepts of causality and generalisation so conflated just in our argument of the definition of randomness.
I am at a complete loss to understand how anyone could find anything objectionable in my definition of randomness.
"1 in 300 billion" is just a number you made up. It means nothing.
And, life arose from not-life, somehow. This is called 'abiogenesis'. There is no way I can assume this is 'impossible', even as a thought experiment.
Taking my toys and going home now.
Advance one. Just one. Argue it logically. Demonstrate how a system could possibly 'know' what mutation would be beneficial without running the program.
"Knowledge" is stored in DNA - this knowledge although taken to be a code for reproduction, has the sequential instructions to build and run an organism as well as reproducing it. It "knows" what it has to do. It has the experience of millions of generations programmed into it. I presume you know that the Weismann barrier I believe to not be a barrier at all. Weismann barrier is also not experimentally derived, by the way. Obviously, with enough "knowledge" you can predict the future, roughly. So I am not stating that it magically knows the future, but that the knowledge and ability to reprogram its own DNA (self modifying code) is also part of the code. Also nothing experimentally derived to say that this does not happen. I went through this with you when we were discussing Lamarkism. I did not understand your objections then, except that to be accepted as scientific hypotheses, it needs to fit in with the framework (including parsimony) as it stands, which is fair enough if I was going to submit it as a formal hypothesis that other biologists will take seriously. However, at this point I am trying to get you just to *understand* what my thesis is here, being that you are roughly familiar with marconomics and have been supportive of my other axioms in the past. The Darwinism inherent in biologists philosophy is closely intertwined with the molecular genetics that is experimentally derived and is referred to when explaining the detail of Darwinism, but the reality is that the main assertions of the modern evolutionary synthesis are *not* experimentally derived, but simplifications and generalisations of experiments that are not 100% understood in the context of speciation.
Ok. Randomness and causality.
We have the technology, sort of. Start with the simplest single-felled organism known, but it doesn't really matter. Use a random number generator and a gene splicer to pick a gene at random and either check to see if this mutation is passed on, or ensure that it is passed on. Compare the result to a set of mutations required to get to a viable variant with a mutation that may be useful. Repeat as many times as required to get a feel for how many lethal mutations, bad mutations happen compared to any "good" mutation.
This would be a proof of principle for the *causality* aspect that is lacking in your definition. Each random mutation is passed on and expressed and tested in the real world. Any more complex example has to be assumed to have the same causality arrangement as this simple example, and thus similar probability outputs.
No matter what logic you use to say that a more complex situation has a different probability profile to this simplified situation, it is no longer *experimentally derived* and thus just a guess, or based on the idea that it must be because it works.
We really have little idea what 90% of the genome of most species do, so experimental derivation of this kind is problematic.
Then there is the idea that because irradiative cause mutations are random that whatever genetic differences that mutagenic progeny have is as similarly random as using a random number generator and gene splicer on the genome. Again, this should have a similar lethal/bad/good probability distribution as our original example to have causality demonstrated, but I am suggesting it is a different profile.
My suggestion is that either the mutation generation is nothing like our original example completely random, or that something in advance of expression as a complete organism is "selecting" on the mutations, or more likely both.
Again, I am convinced in the *negative* that randomness and natural selection is not sufficient. I am not sure why I have to prove even one alternative when this one has no causal or experimental basis, and whatever simplified causal probability calculation you can work out experimentally just doesn't add up.
And, life arose from not-life, somehow. This is called 'abiogenesis'. There is no way I can assume this is 'impossible', even as a thought experiment.
I guess I should elaborate, as you obviously can't see what I am talking about. More specifically, that abiogenesis cannot be from non-life to biological life as we know it. My take on it is that life as we don't know it is hiding in plain sight and we cannot perceive it as life because of incompatible time-scale and nature. What would then be even more interesting than hitting ourself against a brick wall with abiogenesis experiments, is to discover the precedent life form that would be the most accessible. You might think that it's nuts, but your description of abiogenesis as described on your video need not be possible. I'm sure if you try hard enough you can do the thought experiment.
Okay, I will try one... more... time. Since you say you did not understand my objections then, I am assuming you still do not.
It "knows" what it has to do is *not* a mechanism. I sent you the Genes book years ago when we were first talking about this in hopes that you could come up with a mechanism for how this 'knowledge' could be stored and applied in terms of how we know genes work. Give me something... anything... concrete, in the language of molecular biology, for how DNA could encode the knowledge and ability to reprogram itself. I am not going to entertain, or understand, any hypothesis that rejects our painstakingly-derived knowledge of how genes work. A viable theory has to *contain and expand* this knowledge, like Newtonian physics is contained in Einsteinian physics.
My principal objection to your model, as I (mis)understand it, is that in a system as informationally complex as living systems, even in an unchanging environment, there is no way to 'know' whether an arbitrary change in a programme will be good or bad, without running the programme. And the environment in which the programme is run is ceaselessly changing, so 'good' or 'bad' will be dependent on an incredibly complex context. You can blame this on a bias towards free-market ideology, if you like, and a life-long hatred of planning, but I am convinced that even if plausible mechanisms exist they would be inefficient and unresponsive in biology, just as they are in economics.
I am not sure if we are arguing at cross purposes or in parallel universes. You are not understanding my objections to standard evolutionary synthesis specifically and how something not experimentally derived like that gets to be the bar that other theories have to jump over in general as something that can be called "science".
Did you understand my objections to your definitions of randomness?
Do you believe that the standard evolutionary synthesis is experimentally derived?
Do you believe that the DNA code is just a code for reproduction and has no other purpose or knowledge?
Do you believe that DNA never runs as a Turing machine?
Is it only me that believes that whether these things are experimentally derived or not is important?
Did you understand my objections to your definitions of randomness?
No, not in the slightest.
Do you believe that the standard evolutionary synthesis is experimentally derived?
I believe, as a synthesis in the historical sciences, that it *cannot* be experimentally *derived*, but that it is 100% consistent with the results of observations of experiments done by nature, and with laboratory experiments of features of the model, meeting the requirement of uniformity. I am not aware of any experimental observations that would necessitate any change to the standard model.
Do you believe that the DNA code is just a code for reproduction and has no other purpose or knowledge?
The DNA code is a mapping of DNA sequences onto protein sequences. Some of the activities of DNA are related to reproduction, but most are related to the day-to-day functioning of the organism. It does not make any sense to speak of the DNA code having 'knowledge'. This is like saying the alphabet can whistle.
Do you believe that DNA never runs as a Turing machine?
DNA never runs as a Turing machine. I cannot imagine any conceivable physical mechanism for how it could act this way.
Is it only me that believes that whether these things are experimentally derived or not is important?
I have thought in some detail before, as you know, about the relationship of experiment to the historical sciences. I believe it is absolutely fundamental that postulated theories of evolution do not deviate from the experimentally verified physico-chemical facts of molecular biology. As I (mis)understand them, your theories do deviate, grievously.
I think we are making some progress. Just have to keep digging and make sure we don't argue full circle. Your definition of randomness appears to have the desired result as part of the definition. I am hearing - "randomness is the type of variation a DNA strand gets that is selected on" ie. That its randomness is beyond question and that nature is doing the experiment with random mutations and selection such that we don't have to experimentally verify that we get the same results with a simulated random mutation that gets decided on by a process humans can control and verify that which is actually happening in speciation makes statistical sense. As an example that we have discussed to death re "Me thinks I am a weasel" . I am not impressed at all by the counter arguments Dawkins and you have given.
The synthesis, since it cannot be experimentally derived, by your own arguments that it does not fit into the cup of truth and thus is poetry, art, imagination. The modern evolutionary synthesis is compelling, thoughtful and is built on consensus, but I insist *it is not science*. To me, it is religion, and religion can be completely consistent with all known science, and yet be impossible.
*I* think we are getting more and more lost in a fog of meaningless drivel. I can't make head or tail of anything you said in that last comment.
Well, I will use smaller sentences and more references.
What is wrong with this definition of randomness?
http://en.m.wikipedia.org/wiki/Randomness
Randomness is often used in statistics to signify well-defined statistical properties. Monte Carlo methods, which rely on random input, are important techniques in science, as, for instance, in computational science.[2]
Why cannot we use well known statistical methods in the simplest of cases as a proof of principle?
My answer is because it makes the modern evolutionary synthesis look stupid.
Scientists discover second, secret DNA code
Rather than concerning itself with proteins, this one instructs the cells on how genes are controlled.
Because we are reverse engineering the function of DNA etc. bottom up, we are not going to see the big picture as how the whole system works as a set of algorithms.
It is like seeing and knowing what all the silicon etched logic gates in a computer do individually, but cannot reverse engineer what the higher level programming language is.
The only problem with the definition you cite is that it is too general, and does not specifically address the processes of interest to biology as my movie does.
I have to ask again, what do you not understand, or what do you find so objectionable, in my definition of a random mutation? As I said recently, I utterly fail to understand, in any way, what your objection is.
1) That is a really rubbish headline. Of course there are binding sites for transcription factors, stop codons, etc., and these are among the things that will make the location of mutations that escape correction trivially non-random, like AT-rich sequences being less stable than GC-rich sequences.
2) I don't agree that life is algorithmic. I think this is a bad idea from computer science that has infected thought in this area. My apologies for the careless repeated use of the phrase 'run the programme'; this is sloppy shorthand for 'see what happens when you make a change to a complex system governed by multiple non-linear differential equations'.
Often in your discussions with me over the years you have used the term "trivially non-random", and it befuddles me. The difference in trivially non -random and random is the difference between being able to win millions of dollars on the roulette wheels that are not level and having the house win all the time. It is this point in particular where by the statistical definition, it can be in no way be called random, and even if all the non-random bits could be traced and shown to be irrelevant to my case, it still is the wrong word.
What, in the definition in my video, do you disagree with or misunderstand?
Yes, random mutations are not 'random' in the true statistical sense. "Lacking any definite plan or prearranged order" and "proceeding, made, or occurring without definite aim, reason, or pattern" are two definitions thrown up for the meaning of 'random' on dictionary.com. That is what I mean by 'random'.
When talking about this being a *scientific*statement* rather than a statement of faith, the true statistical meaning of the word is surely a prerequisite.
Your video makes it a statement of fact that it does lack any definite plan or order etc. but this is not proven so it is assertion based on what? I take it to be parsimony again. I'm sure you will protest, but if it can't be proven, why are you believing it to be science? Before Darwinism "won" over Lamarkism, there was some concept of experiences during life effecting changes to progeny. All tested experiences had no effect on progeny. The Weismann barrier was invented to explain why that would be. The Weismann barrier is not proven either. Activity of reproduction looks the same whether there is a Weismann barrier or not.
I'm just going to summarise what I believe to be the source of our differences.
1) Experimental basis - your mentioning that a synthesis cannot be experimentally derived by definition, is to me an admission that it is not strictly science. You do believe it to be science by a reasoning I do not accept.
2) expanding informatics principles to genetics - genes as a store of information and instructions analogous to a computer algorithm. To me it is obvious and valid - to you (and biologists in general) it is a big no-no.
3) definition of randomness - Ditching a statistically valid definition of random in favour of a statutory functional one makes the synthesis *not falsifiable* in the statistical sense. One should be able to verify that a simulation based on statistical randomness would come to the same probability distribution.
4) Dismissal of trivial non-randomness. You appear to do this equally for biochemical reasons that mutations would happen in certain areas more than others that are not proportional in any way to the function, but also it appears for things like horizontal gene transfer and epigenetics effects. To me it is an admission that random is incorrect even in your functional description. For instance, I think it is as reasonable a synthesis that horizontal gene transfer explains the spread of all beneficial mutations. I do not think that it is the whole story, but the standard evolutionary synthesis crowds out other ideas as if it had been experimentally verified.
When you ask me "explain to me an alternative mechanism", I have an explanation that obviously involves looking at it in an informatics sense, which you are clearly not going to accept the premise of. Explain it in terms of well understood genetic biochemistry, you ask. Well, my perception is like asking to explain how FaceBook would know that I am fat (to send me useful dieting ads), but to explain it all in machine language terms that we can usefully *reverse engineer* from looking at the program's inputs and outputs and physical mechanical properties, most of which appears to be faithful duplication of FaceBook. Is it reasonable or not to assume that the algorithm can learn and target ads appropriately, and that this can be neither random nor preprogrammed from a higher authority?
One more thing. New paradigms in science are *really hard* for scientists whose day to day profession is about teaching and studying under the previous one. Imagine being told to assume the sun is the centre of the solar system, or to work backwards and assume the Earth was. Then pretend you are well versed in Euclidean geometry and told to assume that parallel lines meet twice. Or that the speed of light is the same no matter how fast the relative speed. Abiogenesis is one of those things that you cannot think what the alternative is until you try and assume that the process (as elegantly described in your video about abiogenesis) that envisions living processes from no living processes through a chemical pathway via a Proto-life as being impossible.
Well, new paradigms are not things people come up with for laughs, but appear when there is a clear and present need. I don't think that need has been demonstrated. And I'm not going to try and believe two impossible things before breakfast*, so you may as well stop asking me to do that. Here's a new post for you to hang comments on.
* That is, I am never going to believe that the entirely credible processes that lead chemically from non-life to life are 'impossible'; and I am never going to believe that unexpressed genes can meaningfully store information over geological time.
There IS a clear and present need. Nobody can see the need before the revolution. After the revolution it's "what were we thinking?" I'll give you a heads up about Stretch Theory. There is several lines of evidence on that link that prove stretch theory independently, and stretch theory is a big deal, because an extensive set of modelling ten years ago after the Hartley comet encounter came to the conclusion that it couldn't happen with any reasonable assumption of material properties, rotational and gravitational profile, etc. thus the mission scientists are prone to ignoring the incontrovertible evidence which is hiding in plain sight. I implore you to read some of the link. Andy C Cooper (alias Scute1133) is similarly enthusiastic about stretch theory, and is equally aghast that the scientists have a blind spot to the evidence and the theory.
Post a Comment