My problem with the IDists’ 500 coins question (if you saw 500 coins lying heads up, would you reject the hypothesis that they were fair coins, and had been fairly tossed?) is not that there is anything wrong with concluding that they were not. Indeed, faced with just 50 coins lying heads up, I’d reject that hypothesis with a great deal of confidence.
It’s the inference from that answer of mine is that if, as a “Darwinist” I am prepared to accept that a pattern can be indicative of something other than “chance” (exemplified by a fairly tossed fair coin) then I must logically also sign on to the idea that an Intelligent Agent (as the alternative to “Chance”) must inferrable from such a pattern.
This, I suggest, is profoundly fallacious.
First of all, it assumes that “Chance” is the “null hypothesis” here. It isn’t. Sure, the null hypothesis (fair coins, fairly tossed) is rejected, and, sure, the hypothesized process (fair coins, fairly tossed) is a stochastic process – in other words, the result of any one toss is unknowable before hand (by definition, otherwise it wouldn’t be “fair”), and both the outcome sequence of 500 tosses and the proportion of heads in the outcome sequence is also unknown. What we do know, however, because of the properties of the fair-coin-fairly-tossed process, is the probability distribution, not only of the proportions of heads that the outcome sequence will have, but also of the distribution of runs-of-heads (or tails, but to keep things simple, I’ll stick with heads).
And in fact, I simulated a series of 100,000 such runs (I didn’t go up to the canonical 2^500 runs, for obvious reasons), using MatLab, and here is the outcome:
As you can see from the top plot, the distribution is a beautiful bell curve, and in none of the 100,000 runs do I get anything near even as low as 40% Heads or higher than 60% Heads.
Moreover, I also plotted the average length of runs-of-heads – the average is just over 2.5, and the maximum is less than 10, and the frequency distribution is a lovely descending curve (lower plot).
If therefore, I were to be shown a sequence of 500 Heads and Tails, in which the proportion of Heads was:
- less than, say 40%, OR
- greater than, say 60%, OR
- the average length runs-of-heads was a lot more than 2.5, OR
- the distribution of the proportions was not a nice bell curve, OR
- the distribution of the lengths of runs-of-heads was not a nice descending Poisson like the one in lower plot,
I would also reject the null hypothesis that the process that generated the sequence was “fair coins, fairly tossed”. For example:
This was another simulation. As you can see, the bell curve is pretty well identical to the first, and the proportions of heads are just as we’d expect from fair coins, fairly tossed – but can we conclude it was the result of “fair coins, fairly tossed”? Well, no. Because look at the lower plot – the mean length of runs of heads is 2.5, as before, but the distribution is very odd. There are no runs of heads longer than 5, and all lengths are of pretty well equal frequency. Here is one of these runs, where 1 stands for Heads and 0 stands for tails:
1 0 1 1 1 0 0 1 1 1 1 1 0 0 0 0 0 1 1 1 1 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1 1 0 0 0 0 1 1 1 1 1 0 1 1 1 1 0 0 0 0 1 1 1 1 0 0 1 1 1 1 1 0 0 0 1 1 1 1 0 0 0 1 1 1 1 0 0 0 1 0 0 0 0 0 1 1 1 1 1 0 1 1 1 1 0 0 0 1 1 1 1 0 0 0 0 0 1 1 0 1 1 1 1 1 0 0 0 0 1 0 1 1 0 0 0 0 0 1 1 0 0 0 0 1 1 1 1 1 0 0 1 1 1 1 0 0 0 1 1 0 0 0 1 1 0 0 1 1 0 1 1 1 1 1 0 1 1 0 0 0 1 1 1 1 0 1 1 1 1 1 0 0 0 0 0 1 0 1 0 0 1 1 1 1 0 1 1 1 1 0 0 0 0 1 1 0 0 0 0 0 1 1 0 0 0 1 0 1 1 1 1 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1 0 1 1 1 0 0 0 0 1 1 1 1 0 1 1 1 1 1 0 0 0 0 0 1 1 1 0 0 0 0 1 1 0 0 0 0 1 1 0 1 0 0 0 1 1 1 0 0 0 0 0 1 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1 0 1 1 1 1 1 0 1 1 1 1 0 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 0 0 0 1 1 1 0 0 0 0 1 0 0 0 1 1 1 1 1 0 0 0 0 0 1 1 0 0 0 1 1 0 0 0 0 1 1 1 1 1 0 1 0 0 1 1 0 0 0 1 1 0 0 0 0 1 1 0 0 0 1 1 1 0 0 1 1 1 1 0 0 0 0 0 1 1 0 0 1 1 1 1 1 0 0 0 0 1 1 1 0 1 1 1 0 0 0 0 1 1 1 1 0 0 0 0 1 1 0 0 0 1 1 1 0 0 0 0 1 1 1 1 0 0 1 1 1 0 0 1 0 0 0 0 1 1 1 0 0 0 0 0 1 1 1 1 0 0 1 1 1 1 0 0 0
Would you detect, on looking at it, that it was not the result of “fair coins fairly tossed”? I’d say at first glance, it all looks pretty reasonable. Nor does it conform to any fancy number, like pi in binary. I defy anyone to find a pattern in that run. The reason I so defy you is that it was actually generated by a random process. I had no idea what the sequence was going to be before it was generated, and I’d generated another 99,999 of them before the loop finished. It is the result of a stochastic process, just as the first set were, but this time, the process was different. For this series, instead of randomly choosing the next outcome from an equiprobable “Heads” or “Tails” I randomly selected the length the next run of each toss-type from the values 1 to 5, with an equal probability of each length. So I might get 3 Heads, 2 Tails, 5 Heads, 1 Tail, etc. This means that I got far more runs of 5 Heads than I did the first time, but far fewer (infinitely fewer in fact!) runs of 6 Heads! So ironically, the lack of very long runs of Heads is the very clue that tells you that this series is not the result of the process “fair coins, fairly coins”.
But it IS a “chance” process, in the sense that no intelligent agent is doing the selecting, although in intelligent agent is designing the process itself – but then that is also true of the coin toss.
Now, how about this one?
Prizes for spotting the stochastic process that generated the series!
The serious point here is that by rejecting a the null of a specific stochastic process (fair coins, fairly tossed) we are a) NOT rejecting “chance” (because there are a vast number of possible stochastic processes). “Chance” is not the null; “fair coins, fairly tossed” is.
However, the second fallacy in the “500 coins” story, is that not only are we not rejecting “chance” when we reject “fair coins, fairly tossed”) but nor are we rejecting the only alternative to “Intelligently designed”. We are simply rejecting one specific stochastic process. Many natural processes are stochastic, and the outcomes of some have bell-curve probability distributions, and of others poisson distributions, but still others, neither. For example many natural stochastic processes are homeostatic – the more extreme some parameter becomes, the more likely is the next state to be closer to the mean.
The third fallacy is that there is something magical about “500 bits”. There isn’t. Sure if a p value for data under some null is less than 2^-500 we can reject that null, but if physicists are happy with 5 sigma, so am I, and 5 sigma is only about 2^-23 IIRC (it’s too small for my computer to calculate).
And fourthly, the 500 bits is a phantom anyway. Seth Lloyd computed it as the information capacity of the observable universe, which isn’t the same as the number of possible times you can toss a coin, and in any case, why limit what can happen to the “observable universe”? Do IDers really think that we just happen to be at the dead centre of all that exists? Are they covert geocentrists?
Lastly, I repeat: chance is not a cause. Sure we can use the word informally as in “it was just one of those chance things….” or even “we retained the null, and so can attribute the observed apparent effects to chance…” but only informally, and in those circumstances, “chance” is a stand-in for “things we could not know and did not plan”. If we actually want to falsify some null hypothesis, we need to be far more specific – if we are proposing some stochastic process, we need to put specific parameters on that process, and even then, chance is not the bit that is doing the causing – chance is the part we don’t know, just as I didn’t know when I ran my MatLab script what the outcome of any run would be. The part I DID know was the probability distribution – because I specified the process. When a coin is tossed, it does not fall Heads because of “chance”, but because the toss in question was one that led, by a process of Newtonian mechanics, to the outcome “Heads”. What was “chance” about it is that the tosser didn’t know which, of all possible toss-types, she’d picked. So the selection process was blind, just as mine is in all the above examples.
In other words it was non-intentional. That doesn’t mean it was not designed by an intelligent agent, but nor does it mean that it was.
And if I now choose one of those “chance” coin-toss sequences my script generated, and copy-paste it below, then it isn’t a “chance” sequence any more, is it? Not unless Microsoft has messed up (Heads=”true”, Tails=”false”):
true false true false false false false true true false true true false true true true false true false true true true true true false true false false false true true true true true true true true true false false false true true false true true false false false true false false true true false false false true false true false true true true true true true false false true false true true true true true true false true true true false true false false false false false false false false false false false false false false true false false true false true true true false true true true false true true false false true true true true true false true false true false false true true false true true true true true true true false false true false false true false false true true false true true true true false false true false false false false true false false true true false false true false true true false true false true true true true true true true true true true true true true false true false false true false false false true false false true true true false false true true true false true true true false false false true false true false false true false true false true true false true false true false true false false true true true false true true false false false false true false false true false true true true true true false true true true false false false true false true false true true true false false false false false false false true true false true false true false true true false false false true false true true false true false true true true true false true false false false false true false true true true true false false true false true true true true false true true false false true false true false true true true true true false false false true true true false false false true true true false true false true true true true true true true false false true false true false false true true true false true false true true false false true true true false true true false false false true false false false false true false false true true true true false false true false false true true true true false true false false true false true true false false true true false false true true false true false true true false false false false false false false true false true true false false false true true true false false false true false true false false false true true false true false true true true true false true false false false true false true true true false false false false true false true true false true true true false true true false true false false true false true false true true false false true false true false true false false true true false false
I specified it. But you can’t tell that by looking. You have to ask me.
ETA: if you double click on the images you get a clear version.
ETA2: Here’s another one – any guesses as to the process (again entirely stochastic)? Would you reject the null of “fair coins, fairly tossed”?
Here’s a sample run:
0 1 0 1 0 1 0 1 0 0 1 1 1 1 1 0 1 0 1 1 1 1 0 0 0 1 0 1 1 1 0 1 0 0 0 1 1 0 0 0 1 1 0 0 0 1 0 1 1 0 1 0 0 1 1 0 1 1 0 1 1 0 1 0 1 0 0 1 1 0 0 1 0 0 1 0 0 0 1 1 0 1 1 0 0 0 1 1 1 0 1 0 0 0 1 0 1 1 0 1 1 0 1 1 0 0 0 1 1 1 1 0 0 0 0 0 1 1 0 0 1 0 1 1 1 1 0 0 1 1 0 1 1 0 0 1 0 1 1 1 1 0 0 0 0 1 1 1 1 0 0 0 1 1 1 1 0 1 0 0 0 1 0 1 0 0 1 1 1 0 0 0 0 1 1 0 0 1 0 1 0 1 1 0 0 1 1 1 0 1 0 0 0 1 1 0 1 0 1 0 0 1 0 1 1 1 0 0 1 0 1 0 1 0 0 1 0 1 1 0 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 1 1 0 0 0 1 1 1 0 1 0 0 0 0 1 1 1 0 1 0 0 1 0 1 1 0 0 0 1 1 0 0 1 0 0 0 1 1 1 0 0 1 0 0 1 1 1 1 0 0 0 0 1 1 0 1 0 0 0 1 0 1 1 1 0 0 1 0 0 0 1 0 1 1 0 1 0 1 1 0 1 1 1 0 1 0 1 0 0 0 0 1 1 1 1 1 0 1 0 1 1 0 0 1 1 0 1 1 1 0 1 1 0 0 1 1 0 1 0 1 0 1 0 0 0 1 0 1 0 1 1 1 1 0 0 1 0 1 0 1 1 1 0 1 0 0 1 0 0 1 1 1 0 0 1 1 0 0 1 0 0 0 1 0 0 1 0 1 0 1 1 0 0 0 0 1 1 0 0 0 1 0 1 1 1 0 1 0 1 0 0 1 1 0 1 1 0 0 0 1 0 1 1 0 0 1 1 0 1 0 0 1 0 0 1 1 1 0 1 0 0 1 0 1 0 1 0 1 1 0 1 0 1 1 0 0 1 1 1 0 1 0 0 0 0 1 0 1 1 1 1 0 1 0 0 0 1 1 1 0 0 1 1 0 1
Barry? Sal? William?
ETA3: And here’s another version:
The top plot is the distribution of proportions of Heads.
The second plot is the distribution of runs of Heads
The bottom two plots represent two runs; blue bars represent Heads.
What is the algorithm? Again, it’s completely stochastic.
And one final one:
which I think is pretty awesome! Check out that bimodality!
Homochirality here we come!!!
GUANO
Yes, I am playing Humpty Dumpty, who specified what he meant by words. And when I use the word “chaos” here I am using it as in the math theory. I have not used it in any other sense, and in the sense of math theory, it is absolutely related to evolution. That is my point. Yes, I blame Barry for changing his current definition mid-argument. He should be more like Humpty Dumpty and state his definition and stick to it. That is what I am doing.
And if you understand the math theory of chaos, you should understand my point: evolutionary theory – classic Darwinian theory, conforms absolutely to the math of chaos – it is a feedback system, and in feedback systems, change over time is profoundly non-linear, and leads to all kinds of unpredictable but highly structured patterns, many of which “look designed”. But aren’t.
What? (btw way, your “surprise surprise” comment is skirting near the edges of my rules for this site – whether you actually think I am posting in good faith or not, you must assume that I am) Yes, Behe has frequently defined an IC structure s one that cannot function if any one part is removed – his classic example being the mousetrap. He has since proposed the concept of “IC pathways” – how many non-advantageous steps must precede a novel feature, but that is a different concept – in any case, logically refuted by the AVIDA experiments.
Which is exactly what Pallen and Matzke did.
That might be how a designer would do try to make a car engine from scrap, and it would be very inefficient. But we are not talking about car engines, and we are not talking about making anything from “scrap” – nor are we talking about a designer. We are, if you insist on a mechanics analogy, talking about using logs as rollers to carry your boat to the river, chopping the rollers into shorter lighter cylinders to make them easier to back to the bow every time one emerged from the stern, whittling a hole down the centre of each so you can run a liana through them to tie them down and keep them from marauding roller-stealers, when you aren’t using them, then threading them on a narrower log one day in an idle moment, then noticing that if you put a roller at each end a narrower log, and made two of them, you could simply put your boat between the rollers, resting on what is now an axle, and save yourself the labour of constantly having to pick up the rollers and move them to the front of the boat when they ran out of the back.
In other words, it’s much easier to synthesis a crude but novel thing from existing bits than it is to synthesise a broken thing from broken parts. And that’s what evolution does – if stuff works, it keeps it, even it it isn’t optimal, and doesn’t discard it unless something that works better still comes along. Even if the better-still thing involves losing the not-quite-as-good thing, although, because you can often get more than one thing out of one protein, and you can also get more than one copy of one gene, you don’t even necessarily lose the not-quite-as-good thing.
No, it isn’t actually like that at all. This is a classic example of why arguing by analogy doesn’t work – because a thing in your analogy doesn’t work, it doesn’t mean the analogue doesn’t work – it could just be a crap analogy, as, I submit, it is in this case. Molecules are not “like” bicycle pumps or pillows”. They are far more like pieces of complicatedly magnetic lego, if you must have a macroscopic analogue. And as anyone who has ever fished around in even an ordinary box of mixed lego, you can withdraw some extraordinarily neat widgets that are literally made from bits of things that previously did something but now don’t. And that’s before we’ve even used electric charges or reproduction and natural selection.
Well, no, they don’t. Which I’d say was a bigger problem for ID, unless you want to postulate an imperfect designer – which is fine – maybe there was a imperfect designer. Except that the imperfections of the designs are exactly the imperfections we’d expect from a Darwinian system, not a foresighted system. The recurrent laryngeal nerve, for instance (especially on a giraffe!). The human female birth canal. The total non-appearance of anything resembling a macroscopic wheel (what self-respecting intelligent designer eschews wheels?) – the glorious exception to the rule being the one case where a “wheel” like system is actually possible by stepwise evolution – at the molecular level.
As Pallen and Matzke showed.
Gallienesque
phoodoo,
That actually sounds like the Intelligent Design version of events. The designer took a box of parts and reused some of them. Do you have an alternative ID version?
While the designer was at it you’d have thought she would have fixed some of the “issues”.
http://www.spineuniverse.com/conditions
I find it amazing that a process such as evolution can produce such artefacts.
Were the spine Intelligent Designed, I’d expect much much more! I expect you’ve never had a slipped disk? Tell me about how perfect the spine is while you are screaming in pain eh?
All you have to do to stop it is come up with an alternative that better explains the data. It’s quite simple.
Seems you can’t do that, so perhaps you should reconsider your position?
Except that it was ME who first mentioned chaos as a description of the initial conditions which Darwinain evolution proposes as the basis for all novel functioning, so why in the world would we switch to discussing mathematical non-linear “chaos theory” when I am talking about the condition of chaos? As in a total lack of organization or order? I am not making up the word chaos, fool.
She doesn’t also get to play humpty dumpty with my words too just because she always does so with her own.
Is the environment
A) Chaotic
B) Mostly similar to how it was a moment ago.
Think carefully….
Perhaps you should write an OP then, then it’s all your words and yours alone.
You keep using that word. I do not think it means what you think it means.
Here.
Please demonstrate that you are discussing this topic in good faith by taking the short amount of time required to read the six or seven page paper and addressing what the authors actually say rather than what you would like them to say.
Nothing is perfect. Everything has problems.
At work the people who are free with the problems but rare with the solutions don’t often end up working on my team. I can pick holes in my own idea quite well enough thanks! That’s the easy part and people who are impressed with their own ability to do that are, well, not rare.
Solutions on the other hand….
Really? What specific hypothesis for “teleological formation” are you proposing? What predictions does it entail? How would one distinguish it from known evolutionary mechanisms?
You do realize that there is not just one bacterial flagellum, don’t you? And you do realize that “random, fortuitous forward progression” is not an accurate description of known evolutionary mechanisms?
Actually, I suspect the answer tot both of those questions is “no.”
Attention phoodoo
As you are new here and Oleg has responded to your comment, I’ll just point out that calling another commenter “fool” is not acceptable. Please read the rules.
Don’t write more stupidly than you have to, Phoodoo. No one here believes any such thing, unless you’re the one exception who believes it, and if so, then your beliefs about how evolution worked are just plain wrong.
I cannot sit without assistance (although fortunately I can still stand, and walk). Yes indeed, ask me if I find it “amazing” that my spinal vertebrae each have the “exact shape” to “stack up” the way they do!
I would definitely prefer some other shape that would allow me to sit – never mind the somersaults – but your precious Interior Decorator didn’t see fit to give me that option.
All the evidence we have is that the process of evolution gave rise to a good-enough-for-further-reproduction option, given the available genetic precursors and the chemical/physical constraints on vertebrate bone growth. Never a perfect one as you claim it should be (according to your idea of what would have happened if evolution were true; nor according to your idea of what perfection your Interior Decorator would have imbued us with). I’m glad it’s not worse, I’m glad it’s as functional as it is, but this crap of yours is in no way true:
Not on your life, Phoodoo. Not on my life!
I think it’s quite clear that phoodoo believes that ID and evolution result in different outcomes.
Perhaps phoodoo would care to propose an experiment that in principle could differentiate between the two?
What do you say phoodoo?
In that case, please read the cited papers, by Pallen and Matzke, 2006 and Liu and Ochman, 2007
You certainly won’t see the case being made if you won’t read the papers where the case is made!
OK, good.
OK, cool. You are right to distinguish between purely “Darwinian” processes (the tendency for features that help reproduction to be reproduced more frequently, as logic dictates it should) and other processes including “drift”, which is the tendency of traits to increase in probability of propagating the more prevalent they become, which, interestingly, is an analogue of my last two examples in the OP – chance processes, but ones which tend to result ultimately in either “fixation” – all members of a population have the trait/all throws are heads, or extinction – no members of a population have it/no throws are heads, even though neither is intrinsically “favoured” by the program).
But can I ask you to make your request a little clearer: you ask for support for “the claim that Darwinian (not the same as “evolutionary”) processes can plausibly account for the evolution (change over time) of such significant functional structures.” I am not sure what claim you are referring to – drift has been around for a while, and it may well be true that a purely “Darwinian” – in the sense of stepwise change, where each change confers a benefit – could account for such structures. Drift is almost bound to play a role. However, evidence for some precursors that do confer a benefit would certainly increase the probability of an IC structure evolving.
Do you know the Lenski et al AVIDA paper? It is open access here, with supporting materials.
Essentially, they showed, by means of virtual organisms in a virtual environment, that if no feature apart from the most complex, ever conferred a reproductive advantage, the most complex would not evolve. However, provided some simpler features conferred some advantage, it did, reliably, but always by different pathways. Interestingly, those pathways always involved many neutral (non-reproductively advantageous steps) and for the most complex feature, always a sharply deleterious step. And it was always IC by Behe’s definition (would not work at all unless all parts were present)
Even more interestingly, the features were classified by what they did, not the method by which they did it. And the most complex feature took many different forms – in other words, the same function evolved many many times, always via many consecutive non-advantageous steps, always by way of at least one markedly deleterious step, and was always IC.
And yet it always evolved, given a minimum number of simpler advantageous possible features.
Is it acceptable to say someone is making an ass of themselves then?
I think what’s very interesting is the plethora of age related illness.
It’s almost as if evolution stopped “caring” what happens to you after you are unable to reproduce. Like optimizing for long, healthy life after reproduction has ceased is somehow invisible to selection.
Ever have any thoughts on that William? phoodoo? If your intelligent designer is all that, what’s with all the illnesses of age? You’d expect a equal spread of illness, the same were you young or old if ID was in charge. Well, I would anyway. Yet that’s not what we see – many conditions are ruled out from consideration for young people (almost always) due to their age.
So, ID thoughts on that anyone? How does ID explain that better?
Thanks, Alan.
Yes, guys, please vet your own comments before hitting “post comment”. We don’t have many moderators here, and haven’t needed many, and in any case we tend to use a light touch. But that means that there is a big onus on posters to moderate yourselves. Do use the sandbox (in which game rules are considerably relaxed) if you want to feel a little less inhibited).
Thanks all.
Please read the rules:
This is not UD. That’s one of the things I like about it.
No, not in this part of the “game”. But we tend to use a light touch. The principle behind the rules is clear enough, I think, and it’s not a question of the words used – naughty words are no more against the rules than any other – what matters is that you assume other people are posting in good faith, are of normal intelligence, and intend to be taken seriously. Saying that someone is making an ass of themselves would mean that you weren’t taking them seriously, and us thus against the rules.
Even if they are. And I’m sure I frequently do (make an ass of myself, I mean).
Something I used to often do was ask “and when you do read it, let me know what the first point you disagree with is and we can talk about it”.
That might make a good OP. That article. The first objection from an ID supporter. Discuss.
Rinse, repeat till end of paper. The only way to do this is one point at a time, as I fear Bill Nye is about to find out 😛
Please do tell me more about Bill Nye … Sandbox if it’s a distractor for this thread?
In that case I apologise, Phoodoo. I did bring up the word myself in this thread, and I thought I had been the first – and I know I made it clear what I meant.
Anyway, I hope it is now clear what I did mean. If you meant something else, then I misunderstood you.
I have given you OP rights btw, if you want to make a post on this.
No indeed. It is absolutely vital that we both agree on the sense in which we are using the words if communication is to occur and the fallacy of equivocation to be avoided! I hope my meaning is now clear, namely that when I agreed that Darwinian evolution is a chaotic system, I mean that it conforms to the math of “chaotic”, or non-linear, systems, in which there are strong feedback loops.
phoodoo,
Given that the ‘Darwinian’ version necessarily implies a process of genetic descent, the existence of sequence homology is entirely to be expected IF such a descent process was involved. Behe set IC up as a challenge to ‘Darwinism’ – remove a single part and the whole ceases to function, ergo Design since stepwise accumulation of parts is forbidden (or highly improbable, without intermnediate advantage). Yet the evidence argues in favour of precisely that stepwise accumulation and exaptation. When we look at sequence homology across a wide taxon, we find evidence of the co-option of parts of one system for another – in the case of the TTSS, it appears that it evolved from the flagellum, not the other way around. You may prefer to think that the Designer did whatever it needed to do, whenever, which approach is consistent with any and all evidence. Great theory, may it serve you well.
Lizzie,
Fair enough.
phoodoo,
Don’t know why you brought up convergent evolution here, but: No it isn’t. ‘Random’ in the evolutionary context does not mean haphazard, unbiased or equiprobable.
phoodoo,
Good thing no-one proposes that as the sequence, then. Common descent, not convergence.
I only pointed this out, because this was in Olegt’s exact response to me in only his second post to me. I would never start off any conversation by attacking or being insulting to the person rather than the point intentionally, without provocation, just so you know.
To be honest, I am quite sure if one were to do a research study of personal insults directed at persons rather than at the topics discussed, on the internet as a whole, and this site as well, it would be quite lopsided, not in evolutionists favor (see PZ Myers, Coyne, Miller etc as popular examples).
But fair enough, Lizzie posts enough without having to also discuss this topic, so enough on that.
I’m curious WJM. Since you already admitted to being almost totally ignorant of molecular biology and genetics, by what criteria will you judge any proposed evolutionary pathways you are presented as plausible or implausible?
How can you be a fair referee when you don’t even know the basic rules of the game?
That’s why I asked for the criteria on how they established the plausibility of their scenarios.
William J. Murray,
Well, this brings me back to my very first point – you gave no metric of evolutionary ‘implausibility’, so I’m not sure why it is incumbent upon me to provide one either. Nor, as others have noted, do you give any equivalent support to the theory that Design is a plausible alternative to the unguided process.
If I were able to reconstruct the precise mutation-by-mutation sequence, complete with selection coefficients, you might find that ‘plausible’, whereas the simplistic notion “being able to move, albeit badly, is better than not being able to at all” does not have the ring of at least informal plausibility to you? Suit yourself.
The minimal level of detail you would consider satisfactory is one which, by your own admission, you would not be qualified to evaluate! How very Catch-22.
Interesting also that the Designer’s main contribution appears to be the means to screw with us by providing bacteria with a range of related motile, secretory and infective mechanisms. Thanks, Designer. Thanks a bunch.
Perhaps you should just read the relevant literature? After all, the flagellum is the ID icon and if you don’t know then it’s up to you to find out, not expect to be spoonfed the entire thing.
As a starting point you could look at the PandasThumb?
Or, you know, just read that single paper you’ve already been linked to? Unbelievable.
You can offer “scientific evidence” all day long; until you make a good faith effort to assure me that it actually relates to my argument, it’s you that has dropped the ball, not me.
That involves pretty much everything that is currently known about molecular biology and genetics. If you start now and work hard you might learn enough to make an informed decision in a year or so. But that means you’d actually have to read the scientific literature instead of just yelling “literature bluff!: and running away.
William, one really important thing you need to know about scientific methodology is that we have no way of computing the probability that a hypothesis is correct. So it’s not what anyone does, and so you won’t find it in the paper.
There are two kinds of probabilities that ARE computed – the probability of observing data as, or more, extreme under some relevant null; and the posterior probability of a hypothesis being better than some alternative. And we can’t do that with ID until it makes some specific predictions, which is why Dembski favours the former rather than the latter.
So you are asking for something that simply is not part of scientific methodology.
There is no such “metric”. What the paper does (the Pallen and Matzke one) is show that there exist molecular analogues, with known functions, of molecules that make up the flagellum, in other bacterial lineages. In other words, that the genetic sequences exist to produce bits of the flagellum, and that those bits confer benefit. This means, that by definition, there are potential advantageous precursors to the flagellum, and therefore a possible pathway, via Darwinian steps, for the flagellum to have occurred.
To put it schematically, let’s say a flagellum consists of parts A, B, C and D, :
If A is advantageous, then if A appears, then lots of individuals with A will result.
This increases the probability that in one of these, B will occur.
If B is advantageous, lots of individuals with A and B will result.
This increases the probability that in one of these, C will occur.
If C is advantageous, lots of individuals with A, B and C will occur.
This increase the probability that in one of these, D will occur.
And if D occurs, you will have your flagellum.
And the probabilities are like this: the probability of A, B C and D occurring simultaneously in one individual are infinitessimal. So are the probabilities of A them appearing sequentially, if none are advantageous except the four-piece combo.
However, if A, B, C and D all confer benefit, the probabilities work like this:
Let’s say the probability of A occurring in any one individual is .00001.
That means that in 1,000,000 individuals (tiny for a bacterial population) the probability is 1-(1-.00001)^1,000,000, the probability of A occurring at least once nears certainty (0.999954602 by my calcs).
If A is not advantageous, that individual may simply die pffpringless. However, if it is advantageous, it is likely to spawn many descendents. And if the probability of B occurring in any one of those individuals is also .00001, then after a million descendents, all with A, there’s a .999954602 that at least one of them will have B.
That gives an overall probability of A + B of .999954602^2, which is 0.999909207
– still pretty good.
And if A+B is advantageous, then we can do the same again for C and get .999954602^3 for A+B+C, which is 0.999863813
And if A+B+C is advantageous, we can do it for D and get .999954602^4 for A+B+C+D, which is 0.999818422
Again, a near certainty. So, by demonstrating that A, B and C were all advantageous, Pallen and Matzke go a long way to demonstrating that the bacterial flagellum is not only possible, but probable.
Now we don’t necessarily have the actual probabilities for the occurrence of each change, nor of the probabilities that each will spawn a lineage of at least a million, but we can take a stab at a ball park.
And in any case, the case for plausibility, at a minimum, is made, I think. Certainly Behe’s case for impossibility is refuted, I would say (and not only by this – there are at least four other nails in the coffin, as I point out in my more recent post.)
Guys, I’m going on a guano binge. The rules are there for good reasons. Please stick to them. They aren’t hard, and there are plenty of other places on the internet where you can break them to your heart’s content.
I’ll just note that I was assuming William had mind powers, as he himself claimed. It was hard to assume he was posting in good faith, but I did!
William J. Murray,
Like this?
The argument is apparently that science has failed to cross your line between ‘possible’ and ‘plausible’. You appear to require a ‘metric’ that will nudge the ball over that line. Yet we (and, I suspect, you) have no idea what that metric might look like – The Evidence That Convinces WJM. You are happy to accept Behe’s handwavy notion of implausibility due to ignorance of intermediate functions, yet the direct evidence that there were intermediate stages, all coalescing upon a single original protein, is insufficient?
The phylogenetic evidence in the paper reveals serial extension of a much more basic apparatus, which is not consistent with the claim that the flagellum is unevolvable due to current interlocking of parts. Improved motility is a plausible driver for serial enhancement in a ‘Darwinian’ competition. To deny that, one would have to argue that the Designer was aiming for, and sustains against degradation, something pointless: moving bacteria.
According to the rules of this site you’re suppose to assume everyone is posting in good faith. That you keep questioning the integrity of the many people who have offered to educate you and ignoring the data they’ve provided is a clear violation of those rules.
Also, if flagella are such big deal, why are there still flagella-less bacteria?
Hey all you ID “scientists” out there since you’ve dredged up the flagellum again: can anyone tell me how the Design of the flagellum (actually at least three quite different designs) fits in with Meyer’s Darwin’s Doubt claims?
We have no fossilized examples of flagella, they’re just too small. However, what we know of molecular clocks (granted not very accurate) indicates the basic proteins families of the flagella have been around for at least a billion years or so. Did the Designer stop in a billion years ago, do the bacteria job, then come back at 530 million years ago for 10 million years to do the Cambrian body plans? Or is the flagella job more recent?
Science always likes to consider the big picture but ID always seems to come up with these disjointed claims that often are directly contradictory. Any details at all you can offer about the actual timeline, design process, and/or manufacturing mechanism will be appreciated.
Whether or not I could evaluate the technical work/claims made via use of some scientific plausibility metric has nothing to do with whether or not one is even claimed to exist in the first place. If you know of such a metric, direct me to it – if not, let’s agree no such scientific metric exists, and agree that Darwinists simply assume Darwinian processes are plausibly (in any rigorous sense) up to the task of creating a functional bacterial flagellum.
So, Darwinists can neither demonstrate (since the evolution of something like a bacterial flagellum takes so long) their claims (about darwinian processes), nor can they provide a means by which they can be shown in some rigorous sense to be scientifically plausible.
I’m not really sure what the characterizations “random” (mutation) and “natural” (selection) bring to the table that is of any practical use in biology. Can someone enlighten me?
Good, then we do agree that there is no such metric; whether or not the Darwinian explanation rises above bare possibility has not been shown in any significant sense.
Empirically we’ve seen it happen: Lenski’s e-coli experiment. And it happened with a very limited population and timescale (in evolutionary terms)
Also here: http://en.wikipedia.org/wiki/Nylon-eating_bacteria
Science doesn’t do rigorous “plausibility metrics” as part of standard scientific methodology. Science does best inferences based on all the available data, like the phylogenetic data on protein families in that paper about flagella evolution you still refuse to read. You’d understand that if you knew anything at all about how science operates.
Early candidate for understatement of the year.
No. You are the only one who can enlighten you. Reading the scientific literature that has been laid at your feet would be a start.
As I’ve said, William, there is no scientific methodology for quantifying the probability that a theory is correct. This applies to ID as much as to any other theory.
I’m not quite sure whether you are not understanding this point or what – but such a probability won’t be forthcoming, ever, because it’s not computable.
What we do instead is fit models to data, and then use those models to predict new data. If the predictions are good, we claim support for our models. We do not claim that they are “probably true” “possibly true” or even “plausibly” true. We simply quantify their predictive power.
In fact science isn’t actually about “truth” at all, although the fact that our models keep making more and more accurate predictions suggest that they are homing in on some underlying reality. They are simply models of reality whose fit to data can be quantified.
As ID makes no predictions, the fit of ID to data cannot be quantified, ID cannot be compared with scientific theories, and does not count as one. It will do when it actually makes a prediction (note that the prediction needs to be in contrast to the prediction made by the theory it is being compared with – you can’t compare two models that make the same prediction).
ETA: actually you can, but only by using some other metric, such as parsimony, and there is no guarantee that the more parsimonious model will ultimately prove the more fruitful.
No no no!!! We agree we ” have no idea what that metric might look like” FOR YOU. Not that there IS NO such metric for scientists and reasonable persons understanding the science.
We don;t know, and we suspect that you also don’t know, what would be The Evidence That Convinces WJM.
I’m sure you just innocently misread Alan Miller and weren’t trying to quotemine him by leaving off that significant clause when you quoted him.
In any case, the evidence certainly agrees with Darwinian explanations to a more-than-reasonable level of plausibility, not just “possibility”.
You seem to have forgotten who you are talking to. I didn’t ask for evidence, because I am not qualified to parse such evidence. I didn’t asked to be convinced, because you cannot convince me that Darwinian processes are capable of generating a bacterial flagellum. I asked to be directed to a paper that at least claims to provide some kind of rigorous examination of the plausibility (probability) of Darwinian forces to generate a bacterial flagellum.
Matzke’s may claim to make it plausible, but I seen no definition of a probability metric where “bare possibility” and “plausible” (or some synonymous values) are defined; as far as I can tell, he just makes a case that in his personal opinion makes the sequence “plausible”. As far as I can tell, he has drawn no meaningful distinction whatsoever between “possible” and “plausible”. As far as I can tell, all he has done is outline a possible route for Darwinian forces; he has not offered anything beyond that.
Since we cannot observe darwinian processes creating a flagellum, I’m looking for a description of the plausibility of the outlined scenario that is something a bit more probabilistically rigorous than personal opinion or an argument by analogy using “just so” connections of possibly available parts and possibly selectable functions in possibly corresponding environments in possibly corresponding time frames with possibly sufficient upgrading and refinements.