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!!!
No, because I find “random” almost as vague as “chance”, and “natural selection” to be a bit of a snare, as it implies a selecting agent, and there is no such agent, just a system (“nature”). My preferred formulation uses neither: self-replication with heritable variation in reproductive success.
“Self-replication” is clear enough. The variation can be “designed” as it is in GM foods, or itself an evolved part of the process, as in recombination, or simply built in to the inevitable infidelities of copying (shit happens). It doesn’t matter where the variation comes from; more crucially, adaptation will happen whether or not the variation is designed by some genetic engineer to promote reproductive success, or whether it just happens to, although clearly the latter is a nice shortcut.
It’s just built in to the dynamics of self-replication with variation in reproductive success. In fact some of the variation needn’t actually contribute to reproductive success at the time it appears – loads of variants in any population at any given time are reproductively neutral. However, they provide a rich pool of variation should the environment change, to favour, for example, people with paler skin, or longer toenails.
Where did I make this claim?
Great! So we agree, there is no way, nor has there ever been a way, to scientifically establish whether or not Darwinian processes are plausibly capable of generating what they are claimed to generate.
Which means we should abandon the terms “random” in descrbing mutations, and “natural” when characterizing selection – unless, of course, those classifications bring something valuable to the table when discussing evolution/biology?
If so, what?
What’s changed? Up till now you’ve flatly refused to read any linked papers, rather insisting that someone provide quotes here from the papers and argue their case with the quotes. Why would you suddenly ask to “be directed to a paper”? Does “directed” imply something different to you that it does to every other speaker of English?
Why would anyone bother to “direct [you] to a paper” when you’ve so recently assured us you won’t read them when we do direct you to them?
If you have indeed changed your mind just now, please verify that you have, and that you now intend to read papers to which you are directed.
That there exists a possible path is enough to dismiss large parts of the ID argument.
You’ve been listening to KF to much I suspect! What are you looking for a probability for? If all possible point mutations are explored in a population each generation, and a path has been offered for Darwinian forces I’d say the probability ~= 1.
Lenski. Currently the best we’ve got, I believe. Is the ability to eat new food more or less complex then the ability to swim?
WJM: “Since we cannot observe darwinian processes creating a flagellum” – well we might given enough time and bacteria. We do see Darwinian processes creating in E-coli. We don’t see ID forces, because no one can say what they are.
Lizzie,
So we agree: Darwinian evolution (by Darwinian, meaning “natural” selection and “random” mutation) has never been scientifically shown to be plausible and offers nothing useful to the evolutionary/biological sciences? That we currently have no way of rigorously qualifying the sequences involved in the generation of any feature like a bacterial flagellum as plausibly being the product of Darwinian processes?
Further, the belief that Darwinian process **can** (beyond bare possibility) produce a functioning flagellum is a matter of ideological faith (as would be the belief that it cannot), since there’s no known way to make such a determination?
(reads rules again, bites tongue) No, we haven’t.
Second candidate for understatement of the year.
If you’ve already made up your mind why are you here?
You keep asking for something that multiple people have already explained to you isn’t done as part of standard scientific methodology. You’ve had it explained that ‘plausible’ is the best inference from the available evidence as determined by the knowledgeable scientific professionals who study and work on the topic.
Is there a reason you keep ignoring the answers you’ve been given multiple times? Do you think those answers will change if you make the same demands enough times?
Lenski.
Yes, instead let’s bring in lots of telic language instead.
Now, what’s new on the table? What can we now do that was not possible before? All mutations are designed. Why do people have cancer William? That’s my first question in this new world.
You’ve not been discussing evolution/biology. Merely your un-understanding of a caricature of it.
Is the ability to eat a different food more or less complex then the ability to swim William?
So wrong as to be humorous. Every time we use a genetic algorithm to produce a novel design we not only show the plausibility of such processes, we empirically demonstrate them working.
Indeed: http://theskepticalzone.com/wp/?p=3085
How does one “see” that a process is Darwinian?
Nope, we don’t agree.
So you asked for an uncomputable metric. Fine. You won’t get it. For the gazillionth time, that’s not what scientific methodology outputs, so asking for it is no more going to get it for you for evolution than it is for ID.
But that doesn’t mean that what is presented is just the authors’ “personal opinion” (there are two authors btw). You don’t get into Nature for publishing your “personal opinion”, not unless you are phenomenally eminent, anyway, and even then, it won’t appear as a research paper.
You get it by presenting an argument and data, which they did. Their argument was that there are good candidates for precursors to the flagellum that confer reproductive advantage and are therefore selectable. I gave you an example of what that would do to the probabilities. If you have responded to that post, I don’t see it. That’s as close as you’ll get to a “metric”.
Much more importantly it is a logical refutation of Behe’s argument, which is, essentially “what use is half a wing?” The answer is “a lot of use, even if not for flying”. And as long as half a flagellum is advantageous, it doesn’t matter that it isn’t a flagellum. Variants with its half-made state will still tend to increase in prevalence.
Why should he? He isn’t writing to answer your question, he’s writing to demonstrate that there are selectable candidate intermediaries, and therefore no good reason to think the flagellum couldn’t have evolved.
A lot more in fact, including specific testable hypotheses. But yes, that’s what he’s done. He’s shown that Behe’s null should be retained; the observed data are not especially improbable under the null of not-design.
In that case read the frigging paper.
Or at least my post.
Genetic algorithms are teleological. They are intelligently designed to solve certain problems.
It’s self-evident. ha.
Yes, that’s right. And they also exist in nature, solving the problem of “staying alive and reproducing”.
Your point?
In the Lenski LTEE it’s done by examining every single generation for the genetic changes and their effects.
If you mean how can we be sure there aren’t magic invisible ID pixies pushing the triplet codons around to create new “Designed” genetic variations? Darn, you got us there.
We’re back to “intelligent falling” I think.
So we modeled this thing on that other thing that doesn’t even work and yet we get spectacular results. Uh-huh. I can just see John Q Scientist now “Well evolution is a crock of shit but I’ll use its principles anyway for solving different problem types. What, it work incredibly well? Well evolution must still be a crock of shit.”
William,
Perhaps you could describe for me, in your own words, Lenski’s experiment and what it was and what it showed?
I’m sure it would go a great deal towards understanding your position. I.E. where was the ID in that experiment.
And if there is a path, and it can be traversed one step at a time, and every point mutation happens every generation in a single population what do you suppose that means?
They’re also modeled directly on empirically observed natural evolutionary processes. If the processes work in the lab why wouldn’t they work in the real world?
Incorrect. The fact that a metric that you would like to see is not available does not allow one to generalize to the conclusion that a set of mechanisms has not been shown to be plausible.
It has been rigorously demonstrated that mutations are random with respect to fitness. Known evolutionary mechanisms have been observed and studied in situ, in the laboratory, and in computer simulations.
Do you have any similar evidence for the plausibility of an intelligent designer causing the evolution we observe? You could start with some evidence for even the existence of such an entity.
Did I? Funny how many people provided links to papers that would supposedly compute that metric (in some meaningful way) that you now claim is non-computable and not something within the purview of scientific methodology to offer in the first place.
Well, there’s no reason why he should, but given that is what I’m asked for, there’s reason to expect that he does when someone refers me to this paper in response to my question.
If the probability metric is not computable, how the computation is “affected” by any data cannot offer any significant inferences.
No, we do not agree. Darwinian evolution, where by Darwinian I mean by means of self-replication with heritable variation in reproductive success, has been shown multiple times to be not only plausible but nigh inevitable, given those starting conditions, and has been, and remains, one of the most powerful scientific we possess, not least in the elegance of its simplicity. Darwin didn’t know how the variance was generated (and I do not recall him using the word “random” at all – at one point he inclined towards some kind of lamarckian process), and while he used “natural selection” as a metaphor derived from analogy with selective breeding by farmers, it’s a metaphor best avoided in my view, as it implies that variation is separable from reproductive success, and they are two sides of the same coin. We don’t get a novel “variant” which then “natural selection” then “acts” on – what happens is seamless – variants either do better, worse, or the same as their peers in the reproductive stakes, and if they do better, then they will tend to become more prevalent. Which they may do anyway, even if they are neutral (or even slightly deleterious), until one day things change and they turn out to be rather helpful.
“Random” is a word like “chance” that needs to be carefully defined in context or you end up with rampant equivocation and fallacious inferences. Both variation-generation and whether a variant is reproductively successful are BOTH largely stochastic, and they are not even, contrary to some assertions, independent: most variants are much more likely to confer similar reproductive chances on their bearer to those that the parent sequence did, so are not “random” in the sense of being “equiprobable” (like those coin-tosses), but they are also more likely than not to be deleterious in a well-adapted population, and but more likely to be advantageous in a poorly adapted one.
No, because you can’t quantify “plausibility”. You can, however, use an explanatory theory or hypothesis to generate predictions that can be quantified for goodness-of-fit against new data.
And you can also quite simply refute an argument that says that there are no possible pathways by showing that there is at least one. One Black Swan.
No, it isn’t a matter of “faith”. You forget (or perhaps you have not yet understood) that in science we do not “believe” that theories are true. We have explanatory models. We have rival explanatory models. And we favour the one with the most explanatory power. Darwinian evolution has a great deal of explanatory power. We can never the exact pathway by which something evolved (unless it is one of Lenski’s lab bacteria) any more, to quote someone else’s example, we can show the exact pathway by which a bunch of broken rocks came to be lying at the bottom of a cliff. But we can show that it is perfectly consistent with the theory that freeze-thaw cycles plus gravity account for the phenomena – even though it might be the case that intelligently placed a couple of rocks there for some unknown reason.
We accept that Darwinian evolution is a highly predictive theory, that Behe’s example is not a bar to it (by showing both in principle and in practice that IC structures and this one in particular, have a possible evolutionary paths as long as some steps confer reproductive examples) and we have no reason to invoke an additional unknown factor to explain it – our null of non-design is not rejected.
It could still be false. We never prove theories are true in science. But if ID is going to compete as an alternative, it needs to make alternative predictions. But it makes none.
William J. Murray,
Why?
I’ve been following the discussion fairly closely for the past few days and don’t recall any such claim being made. Please cite a few of these “many people” to support your assertion.
Possibly because they did not interpret your request as a literal demand for a metric on “plausibility”. By any account the Pallen and Matzke paper proposes a perfectly plausible series of selectable steps for the flagellum, but it does not measure that plausibility because plausibility is not measurable. However, I have shown you a substitute metric, which you have yet to comment on, although I went to some trouble to do some actual sample calcs.
But you seem to be missing the key point, which ironically is a purely logical one: if someone claims that A is not evolvable because A is IC and therefore cannot have intermediate steps that confer reproductive advantage, and someone else shows that structures exist that would be perfectly good intermediate steps to A because all confer reproductive advantage, than that someone’s argument is refuted. Behe may still be correct, and that a designer stepped in to ensure that e-coli were equipped to kill children more effectively, but there is no reason to assume so – just as, to reference Bertrand Russell – we cannot prove that there is no teapot orbiting the sun out near Mars, but nor do we have any good reason to postulate that there is.
Because IC ness, on at least 5 counts, turns out not to be a logical bar to the Darwinian argument. .
Amazing. William just dismisses GA’s as “telic” and therefore irrelevant, when challenged on exactly why other then “William says so” it’s ignored.
Why do any of us bother at all? 😛
“Plausible” is not to my knowledge a statistical threshold. Is probability confusing you again, William?
The probability I computed for you is not a measure of “plausibility”. It was a way of showing you that if we could work out the probability of each separate mutation happening in any one individual (which would be difficult, but not impossible to take a stab at) and we knew the probability that each mutation would confer an advantage in the environment around at the time (again, something we’d need to take some guesses at), we could work out how likely four necessary mutations would occur in at least one individual, if each individual with the novel mutation spawned 1 million descendents (not a lot for a bacterium).
And the result came to near-certainty. In other words, if the precursor mutations are reasonably likely to happen somewhere, and produce some advantage, and the environment is such that a flagellum would come in handy, then flagella will evolve.
And if I’m right, then flagella probably evolved separately in different lineages. If I’ve been too generous, then possibly only once.
Patrick,
Beat me to it! I certainly never made such a claim for the paper I linked.
Have you ever worked out the probability that your automobile is true? Or have you ever worked out the probability that it is at least plausible?
Or do you just drive the automobile because it works?
Obviously a virtual environment with virtual organisms is “intelligently designed”, William. But we aren’t talking about OoL here, we are talking about evolutionary processes, which are premised on the existence of an environment that offers threats and resources and a population of self-replicating organisms. All the GA designers is provide that environment and those organisms and lets them figure out how best to exploit the environment to breed best, using only the mechanism postulated by Darwin.
I’ve seen this argument over and over and I just don’t understand why people don’t get that designing an environment and organisms is the analogue of Darwin’s “few creatures or one” on primordial earth, and that it is what happens next that is operation of Darwinian evolution, once the GA is set up and running, not how it was set up, that is the relevant part of the process.
Typically, I leave mine to run on a spare computer or drive while I get on with other stuff. No intelligent intervention is required (in fact, it usually hinders). Behe proposes that a designer was required to intervene to produce an IC structure. In AVIDA, no lab person had to intervene at all – the system simply reliably turned it up, even though no-one knew in advance how the organism was going to accomplish the function, and even though it accomplished it using different means every time.
thorton! take a methylphenidate or something! Or let off steam at atbc!
But ffs read your posts (then the rules!) before you hit “post comment”!
I love that you used ffs.
William J. Murray,
Heh heh. Using a method from ‘nature’ that doesn’t work! 🙂
It’s from The Creationist Playbook 101. Any empirically observed natural process that can be modeled on a computer proves the original process was Deliberately Designed. That’s why when NOAA models hurricanes to predict their path it proves the Intelligent Hurricane Designer created the real ones. When NASA uses computer modeling to calculate deep space probe trajectories it proves the Magic Gravity Fairies were on the job.
Aww, I’ze just havin’ some funnin’. 🙂
It’s not like you can take any of the ID “arguments” offered here seriously.
Isn’t the “plausibility” of something for you, the reader, to decide at any given moment? The sum total of your knowledge on the subject, fuzzed out to a human scale value? “Unlikely”. “Could be”. “For Sure”.
Would the plausibility of the flagellum evolving be more or less after reading a paper that describes a possible path, I wonder? That’s the reason why the reluctance to actually read the paper linked, I think.
It might make sense.
If you’ve not seen this, it’s great.
http://io9.com/watch-as-these-adorable-robots-evolve-the-ability-to-wa-480989422
It’s eviluition happening right before your eyes! woo.
1 rule, 4 materials.
But the rules say that you must adopt that stance, regardless 🙂
“Assume the other posters are posting in good faith” does not mean you must believe it, any more than “assume a frictionless, spherical mass” means that such a thing must exist.
Oh, that’s awesome! William, watch!
At least WJM is on the right track. In a sense there is a natural function that drives the complexity of evolution, but it the lacks foresight WJM wishes to find in a hypothetical genetic algorithm. Human beings lack foresight in regards to the second coming of Christ. Are the choices I make speeding or stalling the foretold? Yet despite our ignorance and our free will, there is nothing we can do to alter its approach. So too evolution and the advent of life in God’s image.
The universe has freedom. God has foresight.
I recently had a discussion about erosion with a YECcer while train traveling through a long canyon that was clearly still eroding in present times — there were rockslide-warning cables for dozens of miles along the tracks. Now even a delusional YECcer will not deny that “unintelligent” erosion does occur; that is, that not every chunk of rock and particle of dust that slides off the cliff does so at that exact moment/exact spot at the specific order of their Lord Jehovah (nor as the work of Satan, either). Even a delusional YECcer will agree that gravity works steadily and without constant divine direction as a natural force, just as the scientists say it does.
What’s ironic is that it was the christian geologists, actually measuring areas of erosion and sedimentation in 18th century Europe, who made it impossible to reconcile reality with a YEC-bible interpretation — not unless they assume their Lord is a trickster god, fooling reasonable observers with a false appearance of ancient process. “Unintelligent geology” is far more of a direct threat to their literal faith than “unintelligent design” aka evolution ever could be. And yet, for some reason, the modern-day YECcers don’t try to force US schools to teach “intelligent falling” in science classes. I think it’s funny that the IDist gang try to gin up “teach the controversy” only in the realm of evolution. (Although admittedly they do sometimes cast aspersions on the known rates of sedimentation or the atomic basis for radioactive dating etc where geology supports the “evilutionists”.)
Why? I guess it’s easier for the rubes to get riled up by “I ain’t no monkey” than by “that rockslide what crushed Billy’s car were sent by the Lord as a warning to stop drinkin’ and hangin’ around loose wimmen. Thank god Billy’s okay, though”
WJM – the paper I linked, incidentally, isn’t about inferring a pathway by some series of ‘plausible opinion gates’, but is about applying statistical tests to a broad range of flagellum genes from across a wide range of taxa. These tests are objective, and give, with quoted confidence levels (ie a probabilistic measure of likelihood), a sequence of acquisition of parts, the source of any exapted genes, and the direction of transfer. They don’t, of course, give selection coefficients for the amendments, nor every step of the transition, as data is inevitably lost by the very processes of evolution and extinction.
Wouldn’t want you running away with any other idea, such as the caricature you proferred above. You could of course both satisfy your intellectual curiosity and argue more cogently about its contents by troubling to read it.
hotshoe,
So that’s why my cars keep getting mangled.
My teenage son recently found out that his girlfriend has never been punished since she never breaks any rules her parents lay down. His bemused question was “How do you know where the line is, then? I know EXACTLY where the line is!”
Reminds me of how some shade tree mechanics install a cylinder head without knowing the proper torque value:
“Keep tightening the bolt until you feel it get loose again, then back out 1/4 turn”. 😉
Lizzie,
According to Liu & Ochman, there are at least 2 different systems, only one of which they investigated. One species, Vibrio parahaemolyticus, even has 2 independently-coded systems in the one cell, running off different drive mechanisms but sharing the same directional stimulus system. Now there‘s an evolutionary puzzle!