I am working on a series of tutorials to cover the basics of Intelligent Design, especially the mathematics of it. This is my tutorial on Specified Complexity, and I would appreciate any thoughtful criticism of it.
Note that I am specifically requesting criticisms on the content of the video itself, not on applications of the concept that are outside the bounds of the video. This is both to help me (I’m trying to improve my presentation of ID) and to help clarify the conversation (is the criticism of *this* information or of some *other* information).
Can I ask members to consider the rules. I completely understand when one feels very strongly on an issue, one’s strength of language tends to increase too. I hesitate to move comments that are largely substantive but contain a line or two that break rules. But I remind you that calling others liars, or
imputingimpugning* their intelligence is against the rules.Please take a moment to review your comment before pressing “post”.
*H/T Glen
Oh, I won’t be imputing intelligence to several commenters.
[impugning?]
Sorry, don’t want to be pedantic, I just liked the idea that we shouldn’t impute intelligence to certain participants. Not a problem.
Glen Davidson
Remember we can observe these processes in real time in the lab in organism with rapid reproduction rates, such as bacteria. Richard Lenski’s LTEE or the MEGA Petri dish
No indeed. Variation has no need to be other than slight differences. Selection is there to sort them.
Don’t really understand your objection. Consider for a moment the niche as the designing element. Take Blue whales. Let’s suppose there is an advantage in being able to sustain longer dives without needing a breath. Genetic variation in the physiology that affects that ability will come under selection.
Whether there is “teleology” is a non-disprovable question. And anyway, there is no need for the theist to propose the hand of God intervening at that point. Were I a theist, I would propose that God designs the niche when she designs the universe – the niche then designs the organism for free – et voilà 🙂
If the mutation rate (together with the other agents of variation, horizontal gene transfer, the additional processes of meiosis, crossing over etc in sexual reproduction) is sufficient, we don’t need to find anything extra.
“Front loading” has already been mentioned. You seem to be drawn to the idea. It lacks much compared to the explantory power and simplicity of the modern evolutionary theory. ToE puts the organism in its niche by selection and adaptation. When does “front-loading” know when to kick in. Where is the information stored prior to being needed? I don’t need to explain why Blue whales are found in the pelagic ocean rather than the Sahara.
Tell me a bit more about how ID deals with the diversity of life we see now and in the past.
We don’t need to merely “presume” that the ToE is a good explanation. We can test it. We can compare it with data, observation and experiment.
Some or most of what you say here may be correct, depending on interpretation but I don’t see how it undermines ToE or acts as support for an ID hypothesis.
I don’t think so. There only needs to be enough variation for evolution to happen.
Since no one here can seem to come up with an operational definition of intelligence why don’t we all just agree to agree that it does not exist?
How little variation is enough variation for evolution to happen?
GlenDavidson,
Good catch! I’ll edit it and impute you as instigator!
Mung,
Enough of course! How to establish what is enough is perhaps a more difficult question.
That’s always been the ID movement’s problem. An inability to define terms and then go on to state clear, testable hypotheses.
And just like that, Intelligent Design becomes Retarded Design
Heh.
Exactly so. As we’ve just seen johnnyb is perfectly aware that there are groups out there who have plenty of money, empty journals and are willing to fund ID research. But for whatever reason he prefers to offer up excuses and reasons why everyone else should do his work.
Not too lazy but too scared I think. Of what you’ll find if you actually got your ideas tested.
You just construct a path back from your desired conclusion to the facts and then claim that is the only path available.
How’s that working out for you?
johnnyb,
On your site you note you are currently researching:
Well, I know about design patterns. Others here understand cellular biology. Shall we talk about that? I mention this because many ID supporters claim that what we observe in the cell and DNA (multi-layered coded information or somesuch) could only be the creation of a designer like us due to the complexity and interrelatedness of all the moving parts. Change one thing somewhere and untold seemingly unrelated things break. Biology is messy like that.
And yet that’s totally unlike any good professional codebase I’ve ever worked on. Simplicity and minimisation of side-effects are what we see in modern programming. Totally unlike what we see in DNA/biology.
So, can you make the case? I suspect that it’s based on something like the observer pattern being something like how cells communicate or something, not very interesting.
What’s the ETA on the research out of interest?
I’ve always thought that would be an interesting topic to explore. Are you sure you want to be seen contributing to ID research?
Just learned a new word did you?
Unlike intelligence, retardness exists. Want an argument from analogy? I know you guys like those
you mean RD research?
Interesting that when johnnyb was presented with tough critical questions about his “specified complexity” claims he slipped right back into whiny YEC mode. 🙂
Thus Rumraket has chosen the third option, there is no such this as fit and unfit, only more fit and less fit. So you see, everything is fit, but its relative.
And then how do we decide the relative fitness? By deciding if something is alive or dead. If its dead its less fit. How much less fit? Exactly one less fit.
So everything is fit, until its not fit. You start off as 0, then become 1 when you are alive, then return to zero when you die.
Ah, but no, no, the clever evolutionist will say, its not just about being alive, you also get credit for reproducing. But we can’t score it based on how many babies you ACTUALLY have, because amongst other things that would leave them with the illogical conundrum of claiming that Stephen Hawking is more than 3 times more fit than the high school football star who is killed in a drunk driven accident before they can reproduce. So it must be about POTENTIAL rather than actual.
But this is getting confusing, how do we decide who has potential and who doesn’t? Well, simple, you wait until they die, then they either had the potential or they didn’t. 1 or 0. 0 or 1.
A ha, now I gotcha, the evolutionist proudly claims check. Some organisms are sterile, so they never had the potential to begin with, they must be 1-1, even when they are still alive! And what about working bees born without gonads? They must be 1-1-1!! After they die they are minus 2 fitness! Is it possible to get any less fit than a worker bee born without gonads? Well yes actually. A worker bee born without gonads that was never born to begin with but is already dead!! THE ULTIMATE UNFIT BEING. -3 !!!
Oops, I think this isn’t working out like Rumraket had planned.
(edit-I used the word unfit, sorry Rumraket. To be more precise scientifically Rumraket insists we use the term, the Ultimate LESS fit being, because everything is fit, some just less. But then again they never were so…oh,nevermind)
Fitness is a measurement that is applicable before reproduction takes place. It’s the probability your current form will survive and reproduce. An organism with a fitness of 0.6 has twice as much chance of reproducing as one with a fitness of 0.3.
Think of it as a batting average. A .300 hitter has twice the chance of getting a hit than a .150 hitter. That doesn’t mean the .150 hitter won’t get a hit sometimes or the .300 hitter won’t make an out sometimes. And like batting averages, they can’t be computed until after the fact.
Alan – you are, as many do, conflating evolution with Darwinian evolution. Do you know that these mutations occur haphazardly? I’ve just pointed out numerous mechanisms for directed variation. Are you sure that Lenski’s lab is showing haphazard variation, and not teleonomically-guided variation? By what means do you make this determination?
But what if all of the variations within reaching distance are less capable? Selection will happen, but it won’t drive towards adaptation. You are assuming that the haphazard mutations will do this, but there is no evidence of this. When genomes hit their target, we usually later find there was a teleonomic mechanism. This usually takes about 10+ years to do, but that is usually what happens. It takes about 10-20 experiments to show that there was a mechanism behind a mutation to sufficient standards to be published, but it takes zero to make the claim that the mutation was haphazard.
Teleonomy (a subset of teleology, which has been my focus this thread) is certainly a question for which we can find answers.
It is simple. It just doesn’t work. Front-loading knows when to kick in by whatever mechanism is available. That’s the goal of biology – to discover such mechanisms. We know the immune system does this with new antigens. Why is it a stretch to think that organisms do this in other situations? Why is haphazard mutations, for which we have no expectation of working, a better explanation than a teleonomic one, which would give an expectation to its working?
Since we’re just supposing things, let us suppose there is no advantage in being able to sustain longer dives without needing a breath. Does it follow that the ability won’t arise? Whale we’re supposing things, let’s suppose that the ability was present all along, rather than assuming it arose gradually, as you are doing.
This is evolutionary theory, therefore we can suppose anything.
It just happened, that’s all. *poof*!
I love evolution, but story-telling is not the same as science.
Adapa,
What do you think that is based on Adapa? On surviving and reproducing!
If you have survived and reproduced, you will probably survive and reproduce.
I love it when evolutionists try to compare it to a batting average (keiths used to do this). You know what your batting average is based on? On your batting average, ha. Its not based on how fast your hands are, how skilled your hand-eye co-ordination is, how good you are at reading pitching, or even if you can stand up or not. Its based on your PAST performance!
This is precisely what fitness measurements do. They take past performance and say, well this survived this much, I guess it will still survive this much. What is, is.
And to make matters worse, it would be like in baseball if you were allowed to get on base because fans voted for you to be on base. Or because you found a get on base ticket in the gutter. Or if you were standing and some wind moved the base to where you were. Or if they made ten more bases. Or if you got really drunk and fell on a base. Or if the base got really drunk and fell on you.
Baseball with no rules whatsoever.
Let’s call this Patrick’s theory of evolution. Just another in a long line of theories of evolution.
Yes. You’re equivocating. Rumraket already told us what fitness means.
That’s OK. You keep confusing your YECkie cartoon version of evolution with actual scientific evolution.
So it’s an after the fact thing. Right?
Then where’s your evidence for this front-loading? We have sequenced the genomes of extinct species back to 700,000 years. Feel free to show us exactly where in their genomes this “front loading” occurred. Your fact-free claims are more like Creationist pants loading.
Bullshit.
One standard definition says that evolution occurs precisely when there is a change in the gene frequencies found in a population.
– Elliott Sober
It follows, then, that at a minimum, genes are required.
Beyond that, as long as some “imperfect replicators” are producing more copies than other “imperfect replicators” that ought to qualify as ‘evolution’ as you’re using the term. The environment doesn’t have to cause the differential reproductive success. That’s just PatrickEvolution.
Perhaps you’re not the best spokesperson for evolution here at TSZ.
Mung,
It seems you are unable to read for comprehension. Here is the rest of that comment:
As such I would be “contributing” to ID research insofar as I would be disconfirming ID, as noted.
That you are unable to represent my position accurately indicates that I won’t be exploring any such topic with you. It appears you are unable to understand other peoples position.
Mung,
In the same sense that the relative performance of two investments is an after the fact thing. You can’t measure things till they happen. But the differential between two things must take some value. It can’t not. Zero is a value.
Allan Miller,
Then its not the potential for reproduction, its reproduction.
And it also wouldn’t matter if what you reproduce can itself reproduce, because the definition you are using now is fitness equals reproducing.
phoodoo,
Certainly, actual reproduction is necessary for evolution to occur.
Absolute fitness is the mean offspring number of carriers of an allele. That mean includes carriers who do not reproduce at all, as well as those with 1, 2, 3, etc. It would also be contributed to by descendants of a lineage, if they have the allele. So wrong on both counts.
So the mighty intelligent design creationism movement is now reduced to “Can you prove our god isn’t involved?” That’s refreshingly honest, albeit utterly uncompelling.
What are you talking about? When has such a mechanism ever been identified in biological evolution?
I’m reminded of a joke Louis CK’s daughter told him:
Daughter: Who told the gorilla that he couldn’t go to the ballet?
Louis CK: I don’t know, who?
Daughter: Just the people who are in charge of that decision.
It’s cute when a little girl says it. Not so much coming from supposedly serious adults.
Genes, that are replicated imperfectly.
Do you even read other people’s comments for comprehension or is primitive pattern matching all you can afford in your rush to respond? You can relax — I’m pretty sure you’ve got the highest volume of the year award sewn up.
Yes. I comprehended that you failed to mention genes in your definition. And you acknowledge that you failed to mention genes.
I consider that progress.
But you failed to also acknowledge my second point. What is required is that the frequencies change, regardless of whether or not the environment is in some sense a cause of that.
Or you could use some other definition of evolution.
Now what do these genes code for, in your imperfect replicators? If they don’t, then you’re not really talking about genes and we’re right back to square one.
Don’t post nonsense. Then I won’t see it or even think of responding to it.
I’m happy to confirm that I understand evolutionary processes as guided – designed*, if you like. I have identified my designer as the environment – the niche. Variation can arise spontaneously and stochastically, just as long as there is enough (and not too much!) I don’t think you have demonstrated any mechanism for directed variation.
Occam’s razor. If a mutation occurs, why should we need to assume a guiding hand when there is already one in the selection process? You are perhaps falling into the trap that I often hear of – that beneficial alleles are rare in sequence space. Granted the theoretical possible number of DNA sequences is vast and increases exponentially with length. But who knows what beneficial sequences lie undiscovered. As yet we have no way to predict whether any particular DNA sequence might have biological activity.
This makes no sense to me. Are you saying there is experimental evidence that (some) mutations are not random? Do please elaborate.
Not yet though, unless you have some astounding news for me.
That’s handwaving, Jon. Per ToE the organism (or more correctly, the population) is shaped by the niche (yes, there’s feedback, Neil 🙂 ) One explanation for the adaptation and why, where, when and how it happened. If “ID” can’t explain the lockstep between niche and organism that we see, it’s not explaining anything.
I’m not standing in anyone’s way. If you or anyone can come up with a better theory than the ToE and find the supporting evidence for it, then ToE will be discardd for the better explanation.
I’m not sure that’s right. I’m no expert but a quick glance at Wikipedia on our current understanding doesn’t indicate “front loading”.
It’s half the explanation. Selection is not random.
[*design is a well-understood and useful verb but it’s a shame it has been co-opted to such a misleading rôle by the ID movement.]
+1 What Alan said.
johnnyb,
OK, as I promised I was going to post more on the video. Thank you first of all for you efforts to put it together. The presentation was nice, but as you know, I’ve vigorously objected to using Specified Complexity at all for defending and promoting ID and creation, although ironically I still have support for the explanatory filter.
You are correct to point out taking the log of probabilities makes them easier to add, but I don’t think that is good enough to incorporate information theory into ID.
Dembski used a lot of Shannon concepts, but the problem is the Shannon’s work was primarily for communication engineering where there were senders/receivers or broacasters/observers that had some prior understanding about what constituted symbols and what the a priori probabilies were and what is meant by “information”. When I have a computer read a disk, there is a convention of how to count the bits. This doesn’t work for a random artifact like mount Rushmore!
Much of design theory does not compute probabilities using algorithmic complexity! This is brutally evident in trying to infer design from man-made objects or God-made biochemical systems.
And when one tries to used algorithmic complexity to compute probabilities, many times a simple variant of the binomial distribution will suffice. There is no need to try to merge the design inference with information theory, and I argued obliquely in my AM-NAT paper that it actually adds needless confusion.
The better way to express the 100% heads concept being improbable is via the binomial distribution. One can even approximate the deviation of 100% heads from the expected value of 50% in terms of standard deviations. The benefit of this is this comes right out of sophomore level probability statistics books in college and it is beyond reproach. I illustrated how to argue that 100% heads is not the result of chance, and it resulted in a year long argument between TSZ and UD. Here are some of the exchanges:
That said, Jeff Shallit agreed with your approach using Kolmogorov, but I find it like using sledge hammer to kill a fly:
http://recursed.blogspot.de/2013/06/confusion-everywhere.html
The binomial distribution works also for the non-equiprobable distribution (unfair coins) or some situations of loaded dice. It’s a little hard to use binomial for dice since dice have 6 outcomes….you’ll have to do a few kluges.
Below is a figure from my AM NAT paper. It would be extremely convoluted to argue that the dominos and cards house and lock and key are designed based on information theory, furthermore, calculating the probabilities does not involve information theory at all but rather classical physics. If one uses information theory, it would be after all the hard physics calculations were done, and then information theory would be just a superfluous, gratuitous add-on that doesn’t add clarity.
I’ve talked to a college pre-med Christian evolutionist, and after one hour of conversation he became a creationist. There was no need to go into specified complexity. It’s fine as an academic exercise, however.
I’ll also add, an example of how a 1st rate paleontologist became convinced of ID is Gunter Bechley. His 3 hour presentation followed by a Q&A of why he believes now in ID after being an atheist Darwinist most of his life didn’t include anything about Specified Complexity in detail in his presentation.
Finally, I think one metric that is almost useful for ID is the complexity metrics of software based on the number of decisions in algorithm (if, else, switch/case, etc.). The number of dependencies (like say the number necessary nucleotides, or characters in a password) is a good approach to calculating probabilities. But again, one can express all this without information theory.
Otherwise I think you gave the presentation about as well as possible and it is consistent with much of the accepted ID literature out there. I don’t think what you said was necessarily wrong, but just not the way I would teach ID or probability calculations relevant to ID.