TSZ has made much ado about P(T|H), a conditional probability based on a materialistic hypothesis. They don’t seem to realize that H pertains to their position and that H cannot be had means their position is untestable. The only reason the conditional probability exists in the first place is due to the fact that the claims of evolutionists cannot be directly tested in a lab. If their claims could be directly tested then there wouldn’t be any need for a conditional probability.
If P(T|H) cannot be calculated it is due to the failure of evolutionists to provide H and their failure to find experimental evidence to support their claims.
I know what the complaints are going to be- “It is Dembski’s metric”- but yet it is in relation to your position and it wouldn’t exist if you actually had something that could be scientifically tested.
Allan Miller,
Except, of course, for any form of ID that operates in a manner that is indistinguishable from the processes of drift/selection of random variation. How would you go about testing indistinguishable differences?
colewd,
That appears to be quite the claim. Can you back it up? DNA did nothing to usurp evolution – quite the reverse.
Heredity via semiconservative template-copied nucleotide sequences finalised the work of unifying Darwin and Mendel. I am at a loss to see how this undermines evolution.
Allan Miller,
The problem was first mentioned in Yockey’s 1977 paper where he stated that the probability of life was unknowable based to the sequential space of proteins. The current evolutionary mechanisms start with random change. Sequences and random change are like oil and water. When you start up your computer you use a password which is a sequence. It is designed to block random search. It can do this with just 10 characters. Multi protein complexes have thousands of characters. The amount of space that needs to be searched in a sequence is defined as s=n^p
Where s=total search space n=possible letters p=length of the sequence. This problem makes the current mechanisms i.e. RMNS or neutral theory highly unlikely as causes and makes intelligent design a viable competing hypothesis given the inference standard.
He’s still waiting for the train.
https://en.wikipedia.org/wiki/Colon_cleansing
Their entire position is based on probabilities, yet they either cannot or will not do the calculations. Yet they all claim to know that it’s probable. Certainly more probable than ID. Even though they deny certainty too.
P(W|KW)
What the probability of generating the phrase METHINKS IT IS LIKE A WEASEL given the keiths Weasel program? But it’s not guided or directed. Oh no.
That’s about as close as it comes to admitting that you’re assuming your conclusion isn’t it?
Can it be established using the tools of modern science that evolution is unguided and without purpose and that the only tinkerer is blind uncaring indifferent forces and collisions that positively preclude intelligent causation? If so, how?
colewd,
It’s not a problem. Sure, the sequence space of proteins is based on the fact that the alphabet has L ‘letters’, so a string of N of them has a L^N chance of forming spontaneously (if alpha peptide links formed spontaneously, which they don’t).
Who says life originated in that way?
What’s L and N at the origin?
Was first life even made of protein?
Why this obsession with probability calculations?
And why do you think biogenesis, regardless, is a problem for Darwin’s theory, a theory about evolution but for one idle speculation on a ‘warm little pond’?
colewd,
What’s the probability a designer will know how to make a life form?
Mung,
So our entire position is based on probability, yet you have never seen a probability calculation? Yeah, that works!
Can you be sure that those dice you just rolled are really random and not being controlled by invisible ghosts? Does that worry you much?
I’m happy to grant you that I can’t demonstrate what seems to be unguided is actually unguided. It might be and it’s so subtle nobody has ever noticed.
Except you, Mung, you seem to have some idea that evolution is guided beyond that. But, of course, you won’t be saying will you. I can’t think why…
Or is all you are after is getting people to say that no, they can’t prove a negative?
I bet there is a least one person here who could demonstrate that your understanding of such is minimal at best.
Mung,
As far as I can see the rates of mutational change observed in divergent lineages are consistent with the rates of mutation observed experimentally. I don’t assume that; it happens to be the case. So if God is tinkering it is being done so subtly it cannot be distinguished.
Why would I want to exclude positive tinkering? I see no evidence for it, is all. You seem determined to assume that there must be something for your designer to do.
There must be something specific they think the designer is doing in real-time, as you say, otherwise we’d all die.
Mung, Frankie, care to say what that is? Does the designer act all the time or just sometimes? When? How do you know?
Is the designer invoked upon a cell dividing? DNA mutation? Where and when does the designer act?
You’d think they’d have some idea by now. They are 100% convinced that life requires design but can’t quite put their finger on what in particular needs that design….
Allan Miller,
Yockey’s paper is a problem for origin and evolution of life. The sequence problem gets bigger the longer the sequence and the more functionally specific proteins are so this problem gets bigger as evolution moves forward.
Why is this a problem? It does not matter how high up the latter you are, each new step is right there.
While that’s true, address the actual problems, not the strawman 747 junkyard nobody actually thinks is reasonable anyway.
Allan Miller,
The design hypothesis is just an inference from whats being observed in the cell and the fine tuning aspects of cosmology. It is an alternative hypothesis to life forming solely through known natural causes from a common ancestor.
OMagain,
As you start up the ladder the steps get further an further away. Pretty soon you have to travel to the end of the universe to make progress. Sequences are the largest mathematical spaces in the universe. They are the perfect architecture for diversity but terrible for random searches.
I’m not sure I understand you. Do you have a specific example involving biology to hand? Can you demonstrate the exponential runaway with examples?
That sounds like nothing more than an argument from incredulity.
An intelligent designer would not have done it that way.
OMagain,
If we use a cell phone as an analogy, your friends phone number of 10 digits it is one of 10 billion possible numbers you can call. If you increase the number to 13 digits then there are 10 trillion possible numbers 1000 times more. So as proteins get longer the search space gets exponentially bigger. As the proteins get multi functional like nuclear proteins the possible types of solutions get smaller. The formula for sequential space is s=n^p
s=sequential space n=numbers 0 thru 9 p=length of the sequence
for the genome s=4^p
for proteins s=20^p
This is the major issue that made me very skeptical of the theory about a year ago.
Patrick,
It is not an argument, just a statement describing two competing hypothesis.
No, it’s not. There is no scientific hypothesis for intelligent design creationism. “Goddidit” is not an hypothesis.
AND
These two issues are related. Colewd’s argument about the unsearchably large nature of sequence space applies similarly to the configuration space of a decent-sized protein. So if protein sequences cannot be found in the sequence-space without intelligent assistance, then protein conformations cannot be found within the configuration-space without intelligent intervention either. In other words, the ‘designer’ must step in to help every single enterotoxin molecule fold correctly. That’s one extremely busy and malevolent ‘designer’
😉
Alternatively, they aren’t random searches…
Patrick,
You are creating a straw man here. The design argument that competes is simply an inference. The observation of design in nature.
DNA_Jock,
My argument for is for coming clean and saying we really don’t understand this. I think that science builds houses of cards when the inference standard is used.
Sure it is.
What entailments does a hypothesis have that a conjecture doesn’t?
We’re fine with designed searches. But that doesn’t mean intelligent design requires a search algorithm.
I don’t understand the question. But when I look up ‘hypothesis’ a given synonym is ‘conjecture’.
Ah. In the realm of science.
Mung,
If you don’t want to admit that you’re watching the Khan Academy videos, that’s fine, but do it anyway. You won’t learn if you don’t make an effort.
Wouldn’t you like to understand what Dembski was arguing all those years?
Aren’t you tired of depending on the critics to explain Dembski’s argument to you?
colewd:
colewd,
I thought you read Arrival of the Fittest.
No?
If you understand that book, you’ll understand why it is the structure of the fitness landscape, and not its size, that matters.
Well, “unknowable” is a bold claim. I’m not sure what Yockey meant (if he meant it) by relating the probability of life to protein space. If he was referring to origin of life then sure, it’s a field full of theories and sparse in evidence. But RNA world is a supportable explanation for not being concerned with proteins at the point when life on earth got going.
I’m still unsure whether you are talking about life’s origin or its subsequent diversity. Evolutionary theory only explains the latter.
And if evolutionary processes were anything like cracking passwords, you’d have a point against evolutionary theory. “Needs to be searched” is where you go wrong, I think. Bumping in to a better (in the sense of more efficient in exploitation of a niche compared to the competition) is what happens. It’s not really a search. Organisms just live, if they can.
Well it would be, if that was how thing are!
Except there is no “Intelligent Design” hypothesis. I can state the central concept of evolutionary theory in a sentence or two. Nobody has yet produced a coherent testable theory of “Intelligent Design” relating to life on Earth as we observe it.
ETA what => when
It’s not a straw man. You yourself claimed that “It is not an argument, just a statement describing two competing hypothesis.” There is no scientific hypothesis for IDC and no IDCist has quantified how to detect “design in nature” in any measurable way.
Your problem is that, when you get right to the bottom of it, intelligent design is creationism. You don’t have a competing hypothesis or inference, you just have religious beliefs attempting to masquerade as science.
What the structure of the fitness landscape of the Weasel program? I’m guessing it has a big hole in it.
What do you mean by a “big hole”?
Is there anything else intelligent design does not required? I imagine the list is quite lengthy.
How is evolution guided then, by the intelligent designer?
Write some code and find out. You might even learn something.
Patrick,
Again you are creating a straw man to avoid the argument. If your scientific standard is inference the design hypothesis competes. If you limit the evolutionary argument to only testable hypothesis then the design argument is not longer on a level playing field.
OK, then. Give us an untestable “Intelligent Design” hypothesis.
Alan Fox,
When you have a testable mechanism for evolution or limit your description to testability then you can make this argument but with the current standard set originally by Darwin the design hypothesis competes on an equal playing field. Both are inferences based on evidence. If your standard is the scientific method then I agree with you but then modern evolutionary theory is an untested hypothesis with some conflicting evidence.
Modern evolutionary theory is well-supported by multiple lines of evidence. What do [you] consider as conflicting evidence? Just the best example that occurs to you is fine.
ETA you
colewd,
Ok…here is a swag at an ID argument
The observation of the sequential structure of DNA that codes for proteins resembles other structures that are designed by man. Those structures include telephone numbers and the english language. The sequential space of the structures can be defined by the same mathematical equation. s=n^p where s is the sequential space
is the number of nucleotides or amino acids in the sequence and p is the length of the sequence. Therefor design is the best inference for this observation.
I don’t agree. Humans are only just (the last century or so) beginning to design and produce complex molecules. DNA has been around for maybe three billion years.
You may be conflating meaning and content! 🙂
This argument relies on an all-at-once probability which is not what evolutionary theory posits.
But that is indeed untestable. You are resorting to “design” as a default explanation after supposedly eliminating another theory. KF (don’t know if you are familiar with this denizen of Uncommon Descent) would say “every tub must stand on its own bottom”.
colewd,
You wrote words and Shakespeare wrote words, therefore you are Shakespeare
It’s worth noting that living things reproduce. This means that at any given time, there’s generally going to be more than one specimen of any given kind/type/species. So when someone talks about the probability of some Mutation X occurring, it makes sense to ask whether they mean the probability of Mutation X occurring in any one particular critter, or, instead, the probability of Mutation X occurring in any of the N critters of that one particular kind/type/species. Because those two probabilities are distinct figures; they’re related figures, to be sure, but they are not the same figure.
As a concrete example of what I’m talking about here, consider a roulette wheel. Let’s say that this particular wheel has slots for the numbers 1 through 36, plus a 37th slot for 0, and a 38th slot for 00. Let’s further say that this particular wheel has not been gimmicked in any way, so that in any one spin of this wheel, the probability of its coming up 00 is 1/38. Since the probability of its coming up 00 is 1/38, it follows that the probability of is coming up anything but 00 must be 37/38.
In general, we can say that if the probability of some Event E (the wheel coming up 00 on any one spin, in the wheel example) occurring is p (1/38, in the wheel example), the probability of that Event E not occurring (the wheel coming up anything except 00 on any one spin, in the wheel example) must be 1 – p ([1 – 1/38], in the wheel example).
Now let’s say we’ve got 10 ungimmicked roulette wheels, instead of just 1 ungimmicked wheel. If you spin all ten of these wheels, what’s the chance that 00 will come up on any of them? Well, for each wheel, there’s a 37/38 chance of it not coming up 00. And since none of the 10 wheels are gimmicked, each wheel’s spin is completely independent of any of the other wheels’ spins. So the probability of all 10 wheels failing to come up 00, must be (37/38)^10 —which works out to a wee bit over 3/4 (or 0.765916234, if you want a more precise figure). So if you’ve got 10 ungimmicked wheels, and you spin each wheel once, the probability of 00 coming up at least once in that set of 10 wheel-spins is a wee bit under 1/4 (more precisely, 0.234083766).
In general, if some Event E has N independent opportunities to occur, and the probability of Event E’s occurring in any one of those N opportunities is p, the probability that Event E will not occur in any one of those N opportunities is (1 – p)^N. Which means that the probability of Event E occurring on at least one of those N opportunities is 1 – (1-p)^N.
Let’s apply this to biology.
ID-pushers like to make noise about how vanishingly improbable mutations are. Well, maybe so; maybe the probability of some Mutation M occurring in any one critter is, indeed, vanishingly small. But thanks to reproduction, living thinks don’t usually exist as one-of-a-kind specimens. Bacteria exist in astronomical numbers themselves; the last time I checked, the best estimate for the total number of bacteria in Earth’s biosphere is about 10^30.
If the probability of some Mutation M occurring in any one specific bacteria is , the probability of that Mutation M not occurring in any one specific bacteria must be (1 – p). And the probability of that Mutation M not occurring in any one of 10^30 bacteria must then be (1 – p)^(10^30).
If Mutation M is so rare that the probability of its occurring in any one specific bacteria is 10^-50, the probability that this Mutation M will not occur in a group of 10^30 bacteria is (1 – (10^-50))^(10^30). Assuming the calculator at Wolfram Alpha isn’t leading me astray, that probability is almost exactly (0.999 999 999 999 999 999 990). Which means the probability of this Mutation M occurring at least once in that group of 10^30 bacteria, is about 10^-20. Not exactly a large number, in absolute terms, but certainly a larger number than 10^-50. Which means that the probability of this Mutation M occurring at all is rather larger than that bare 10^-50 figure might lead one to believe.
So, to repeat, it makes sense to distinguish between the probability of some Mutation M occurring in one particular critter, and the probability of that Mutation M occurring at all.