I thought about including this in my previous thread, but it has grown so large that I suspect it would be lost in the abyss. If Skeptical Zone readers are interested I’ll write a series of these posts, in which I’ll develop a number of themes concerning why I abandoned evolutionary orthodoxy and became convinced that an inference to design is most reasonable.
As most of you know, I am a classical musician. All great musical compositions have a theme, and the theme of this site is “think it possible that you may be mistaken.” With that theme in mind, might I suggest some skepticism concerning probabilities?
One doesn’t need precise numbers to recognize when proposed probabilities are way of whack. When I was growing up and learning mathematics my dad (a professor of chemical physics) admonished me to always check my calculations to see if they made sense on the surface (in my engineering department we call this “using the beverage out the nose” test). If I punch 87 x 53 into my calculator and get 46,481 I immediately know something is wrong (in this case I hit the 7 key twice by accident) even if I don’t know exactly what is wrong, because the result should be somewhere in the hundreds, not thousands. I don’t need to know exactly what the problem is in order to recognize that the result makes no sense.
I apply this logic to probabilities concerning evolutionary theory. We have some good empirical evidence that it took about 10^20 reproductive events for malaria to evolve chloroquine resistance. It could be that Lucy turned into Lizzie in 3.2 million years by stochastic Darwinian mechanisms filtered by natural selection, but I apply the beverage-out-the-nose test concerning the probabilities. Even given the most generous assumptions (a few hundred thousand generations with a few million individuals in each generation) the probabilistic Lucy–to-Lizzie resources don’t pass the smell test, in my view.
So, I ask Skeptical Zone readers: Is my skepticism unwarranted, and if so, why?
William J. Murray,
But this is as absurd as saying “unless you can demonstrate that invisible pink unicorns are unnecessary to account for an arrangement of matter, don’t make the assertion that invisible pink unicorns don’t exist!” test
William J. Murray,
But this is as absurd as saying “unless you can demonstrate that invisible pink unicorns are unnecessary to account for an arrangement of matter, don’t make the assertion that invisible pink unicorns don’t exist!”
Error
Replace “Darwinists and Atheists” with “creationists” and you’re bang on. Coming up with huge numbers evaluating the probability of a trait that has already evolved evolving in the first place and then using the huge numbers to conclude design is standard ID argumentation. You’ve already demonstrated it.
Matter being thrown together in a single step is a caricature of evolutionary theory as Ian Musgrave has already explained. I’d also point out that the machines that fire baseballs at prespecified locations and speeds are not intelligent, they are machines. Natural selection is a mechanism with a healthy element of determinism (or ‘law’ using EF-speak) which dramatically increases the likelihood of finding targets beyond pure chance.
Finally, given the diversity of cell types present today, I think your assertion that there are only a few ways to assemble one is mistaken.
I’d say that some arrangements of matter, including battleships, and also cacti, require something that reproduces with heritable variance in reproductive success to be involved in its production.
Prof FX Gumby,
Natural selection is just a result and doesn’t do anything.
Yes indeed. As I said above:
Nice to be on the same page 🙂
This leaves you arguing that genetic birth defects are products of design, which comes across as a bit crazy IMO.
Prof FX Gumby,
The existence of alleles demonstrates that There are functional sequences adjacent to other sequences having the same or equivalent function.
No, of course not. I’m claiming that all arrangements of matter can be described as the result of natural mechanisms, as opposed to supernatural mechanisms or the operation of nontestable hypothetical entities. Don’t need those.
No, we have established that some arrangements of matter ARE the results of intelligent agency. We determine this by establishing the history of these arrangements. We simply cannot tell if an object was intelligently designed by an intelligent agent, without background knowledge of the process and the agency.
No, you have it exactly backwards. The default is that no intelligent agency was involved, because proving a negative (no such agency) is not possible. Whoever makes the positive assertion (that such an agency was involved) has the onus of demonstrating that this is so.
This seems one of the key blind spots of ID – there really is no positive hypothesis of ID. Instead, and invariably, the argument is that ID is the default, it’s what is left over after some other explanation is doubted vehemently enough. I don’t need to “prove” that no intelligent, invisible, indetectible, inherently untestable agency carefully arranged those glass shards. You are demanding that your intelligent agency be the default, and only by proving a negative can your default be disputed. Granted, this is a very convenient position for you to take. If science adopted that position, science would cease to exist.
William J. Murray,
No, of course not. I’m claiming that all arrangements of matter can be described as the result of natural mechanisms, as opposed to supernatural mechanisms or the operation of nontestable hypothetical entities. Don’t need those.
No, we have established that some arrangements of matter ARE the results of intelligent agency. We determine this by establishing the history of these arrangements. We simply cannot tell if an object was intelligently designed by an intelligent agent, without background knowledge of the process and the agency.
No, you have it exactly backwards. The default is that no intelligent agency was involved, because proving a negative (no such agency) is not possible. Whoever makes the positive assertion (that such an agency was involved) has the onus of demonstrating that this is so.
This seems one of the key blind spots of ID – there really is no positive hypothesis of ID. Instead, and invariably, the argument is that ID is the default, it’s what is left over after some other explanation is doubted vehemently enough. I don’t need to “prove” that no intelligent, invisible, indetectible, inherently untestable agency carefully arranged those glass shards. You are demanding that your intelligent agency be the default, and only by proving a negative can your default be disputed. Granted, this is a very convenient position for you to take. If science adopted that position, science would cease to exist.
No, of course not. I’m claiming that all arrangements of matter can be described as the result of natural mechanisms, as opposed to supernatural mechanisms or the operation of nontestable hypothetical entities. Don’t need those.
No, we have established that some arrangements of matter ARE the results of intelligent agency. We determine this by establishing the history of these arrangements. We simply cannot tell if an object was intelligently designed by an intelligent agent, without background knowledge of the process and the agency.
No, you have it exactly backwards. The default is that no intelligent agency was involved, because proving a negative (no such agency) is not possible. Whoever makes the positive assertion (that such an agency was involved) has the onus of demonstrating that this is so.
This seems one of the key blind spots of ID – there really is no positive hypothesis of ID. Instead, and invariably, the argument is that ID is the default, it’s what is left over after some other explanation is doubted vehemently enough. I don’t need to “prove” that no intelligent, invisible, indetectible, inherently untestable agency carefully arranged those glass shards. You are demanding that your intelligent agency be the default, and only by proving a negative can your default be disputed. Granted, this is a very convenient position for you to take. If science adopted that position, science would cease to exist.
I apologize for working around the link to the post I’m responding to. Whenever I include that link by clicking on (Quote in reply), I am unable to post. Instead, all I write goes away, and I’m back at the top of the thread. My post is nowhere to be found. If I post it again, I get a duplicate post message.
By trial and error (MANY trials), I’ve found that removing the back-link allows me to post. I don’t know why.
Thanks, Flint, for solving the problem of disappearing posts.
I have posted a few comments today that disappeared (not that they were of much consequence) including one making the point about proving a negative, less eloquently than you have subsequently done!
ETA that worked!
First, the idea that it is not possible to prove a negative is a myth. Many (if not most) negative claims are in fact positive claims; “There is no milk in this jar” is a positive claim about the contents of the jar, and such a negative claim can certainly be supported – that there is no milk in the jar.
Second, you shouldn’t be making claims that cannot be supported.
Third, you’re conflating a default premise with an asserted conclusion. If you’re going to assert that ID was not necessary, you have to support your assertion or withdraw it. If you cannot show your non-ID processes sufficient as a reasonable explanation, then you certainly cannot rule out ID as being potentially necessary.
William Murray:
We don’t need to rule out ID because no-one has yet come up with a testable hypothesis of ID. I don’t need to rule out the existence of invisible pink unicorns either!
So, if we land on a distant planet in another solar system and find markings on a cave wall that seem to symbolically correlate – fairly exactly, with near-perfect circles and a ring around one corresponding to a ringed planet in that solar system – to that planet’s solar system, down to moons and orbits, and an arrow pointing at the circle that would correspond to the position of the planet with the cave, and underneath all of that was a series of markings that appeared to be a map of the cave’s location on the planet (an arrow marking it), and under that markings that appeared to be a map of the local area with the cave it self marked in its correct location as per local geoological landmarks, and if the planet is utterly barren and lifeless, as far as you’re concerned those markings cannot be considered evidence of ID because we have no background knowledge of the process or agency that created the markings?
William J Murray,
You’re not proving a negative, you’re proving the positional relationship between two known things, in this case milk, whose existence has been confirmed by supermarkets, and a jar, whose existence has been confirmed by you.
I didn’t say you need to rule it out; I said that if you are going to assert it isn’t necessary to the explanation, you have to support that assertion by showing the non-ID processes you say are sufficient are, in fact, categorically sufficient. You can’t rule out the involvement of ID, you can only demonstrate that your non-ID processes are either specifically or categorically sufficient.
Until that is demonstrated, one cannot rule it out as potentially being necessary to the explanation.
William:
It might well be evidence of extra-terrestrial life. What is the connection with ID?
Alan Fox,
Why would it be evidence of extra-terrestrial life?
Alan Fox,
Of course we have a testable hypothesis for ID. OTOH your position doesn’t have one.
You seem to have left out the actual testable hypothesis in your response, Joe. Please do provide it.
Now that I’ve recovered from the shock of being ganged up on by both Joe and Elizabeth, I’ll rephrase.
“
Natural selectionThe numerous environmental pressures, including inter- and intraspecific interactions that result in the effect known as ‘natural selection’ form a mechanism with a healthy element of determinism (or ‘law’ using EF-speak) which dramatically increases the likelihood of finding targets beyond pure chance.”Better? I think my essential point still stands that “blind watchmaker” or “pure chance” is not a good characterisation of evolution. Thus, calculating probabilities that some trait would develop from nothing is meaningless.
I think babies and calves confirmed the existence of milk first. The citified youth of today – really now.
Joe G,
Or, you know, you could compare the chimpanzee and human genomes to see where the differences lie. I wonder if anyone’s done that 🙂
We still don’t know. We have a testable hypothesis. We could be sufficiently certain that no testing would usually be required IF we knew that humans were the artists. And let’s face it, the “intelligent designer” is basically an agency that does things in a way that is congruent with human designs.
But hopefully, you realize that even in creating your scenario, much less in interpreting it, you are trundling in whole reams of background knowledge of human designs, human viewpoints, human artistic practices, human cartography, etc. etc. etc. Do you not even realize this? Do you sincerely believe that some hypothetical alien (recall what “alien” MEANS) would follow traditional human practices so faithfully? You are doing what’s called stacking the deck in your favor. Which is not generally considered honest practice.
We still don’t know. We have a testable hypothesis. We could be sufficiently certain that no testing would usually be required IF we knew that humans were the artists. And let’s face it, the “intelligent designer” is basically an agency that does things in a way that is congruent with human designs.
But hopefully, you realize that even in creating your scenario, much less in interpreting it, you are trundling in whole reams of background knowledge of human designs, human viewpoints, human artistic practices, human cartography, etc. etc. etc. Do you not even realize this? Do you sincerely believe that some hypothetical alien (recall what “alien” MEANS) would follow traditional human practices so faithfully? You are doing what’s called stacking the deck in your favor. Which is not generally considered honest practice.
We still don’t know. We have a testable hypothesis. We could be sufficiently certain that no testing would usually be required IF we knew that humans were the artists. And let’s face it, the “intelligent designer” is basically an agency that does things in a way that is congruent with human designs.
But hopefully, you realize that even in creating your scenario, much less in interpreting it, you are trundling in whole reams of background knowledge of human designs, human viewpoints, human artistic practices, human cartography, etc. etc. etc. Do you not even realize this? Do you sincerely believe that some hypothetical alien (recall what “alien” MEANS) would follow traditional human practices so faithfully? You are doing what’s called stacking the deck in your favor. Which is not generally considered honest practice.
With respect, William, it isn’t a “myth” at all. Recall, that we don’t “prove” things in science anyway, we infer them, and it is much harder to infer a negative than a positive for the simple reason that the two things are not symmetrical. To demonstrate that there is milk in the jar, all you have to do is to find a trace of milk. You do not have to do an exhaustive search of the jar. However, to demonstrate that there no milk in the jar, you’d need to do an exhaustive search for any trace of milk. Hence the “one black swan” example. All you need to demonstrate that black swans exist is one black swan. To demonstrate that no black swans exist you have to examine every single swan.
This is why it is far easier to demonstrate that, for example, smoking causes cancer than to demonstrate that vaccines don’t cause autism. No scientific search is exhaustive, so at best we can only say that “if vaccines cause autism, they do so so rarely that we were not able to detect an association after a very large study”.
I agree. Not sure who was.
I agree that science has not ruled out ID. One problem of course, is that ID isn’t falsifiable, so it never will.
However IDists claim to have ruled out non-ID processes. This is a problem for ID.
But if we’re allowed to posit invisible indetectible milk, put there by a Designer who has hidden both Himself and the milk from us, how could we ever establish that there is no milk in the jar?
In general, negative claims fall into two categories – those that can be demonstrated by process of elimination and those that cannot. In ordery understandings, process of elimination can be used to establish there’s no milk in the jar. But it can NOT be used to eliminate every possible explanation for biological change and variation. When no evidence is relevant (which is how ID is “supported”), the number of possible explanations is infinite. Process of elimination can’t work.
And this is why ID proponents have been asked to say something, anything, about ID which can be falsified by test. Assuming as the default invisible, hypothetical, in principle indetectible agencies, and demanding proof that they weren’t involved, is a rather silly approach. Certainly not rational.
Flint,
Alan Fox said:
How like human intelligence non-intelligence may be is entirely irrelevant to the argument here. It might be a lot like human intelligence; it might not be. The question is whether or not a design inference (not “knowledge” that something is designed) should be made in the example I provided, which provides no background knowledge as Mr. Fox said was necessary.
You said:
What is the “testable hypothesis”, and how would it be tested?
Then it’s a good thing nobody asked for any such thing.
Elizabeth, olegt and GilDodgen
Re: “Why don’t you go through these simple calculations step-by-step?”
John C. Sanford invented the “gene gun” and specializes in genetic engineering. He is “looking at the theoretical limits of mutation/selection.”
To automate that process of “simple calculations”, Sanford and colleagues created Mendel’s Accountant to provide quantitative forward population modeling with mutations. They include all the major parameters under full user control with free download of the software. See:
http://mendelsaccountant.info/
May I recommend a post on Mendel’s Accountant to
1) Address these issues of quantitative forward population modeling with mutations.
2) Ask users to see if they can find some combination of “realistic” parameters for which there is a net beneficial trend. By “realistic” I refer to parameters published in journals as to typical ranges.
(Or find a feasible set and get it published.)
This would provide actual hand’s on testing and understanding of NeoDarwinian genetic evolution.
In “Genetic Entropy & the Mystery of the Genome” (Lightning Source Inc, Mar 30, 2008 – 248 pages, ISBN 9780981631608) Sanford provides a popular summary of genetic mutations with an appendix describing the major Darwinian population models.
Following is the description of Mendel’s Accountant. ———————————————————————————————–
About Mendel’s Accountant
You may download the program at http://sourceforge.net/projects/mendelsaccount.
Mendel’s Accountant (MENDEL) is an advanced numerical simulation program for modeling genetic change over time and was developed collaboratively by Sanford, Baumgardner, Brewer, Gibson and ReMine.
MENDEL is a genetic accounting program that allows realistic numerical simulation of the mutation/selection process over time. MENDEL is applicable to either haploid or diploid organisms, having either sexual or clonal reproduction. Each mutation that enters the simulated population is tracked from generation to generation to the end of the experiment – or until that mutation is lost either as a result of selection or random drift. Using a standard personal computer, the MENDEL program can be used to generate and track millions of mutations within a single population.
MENDEL’s input variables include such things as mutation rate, distribution specifications for mutation effects, extent of dominance, mating characteristics, selection method, average fertility, heritability, non-scaling noise, linkage block properties, chromosome number, genome size, population size, population sub-structure, and number of generations.
The MENDEL program outputs, both in tabular and graphic form, provide several types of data including: deleterious and beneficial mutation counts per individual, mean individual fitness as a function of generation count, distribution of mutation effects, and allele frequencies.
MENDEL provides biologists with a new tool for research and teaching, and allows for the modeling of complex biological scenarios that would have previously been impossible.
You can
Run the program
View screen shots
Download Mendel’s Accountant from Sourceforge (see download instructions here
Read about the authors
Download the User’s Manual and Linux How-to (requires Adobe Reader)
Join the discussion group
References
J. Sanford, J. Baumgardner, W. Brewer, P. Gibson, and W. Remine. Mendel’s Accountant: A biologically realistic forward-time population genetics program. SCPE. 8(2), July 2007, pp. 147-165.
J. Sanford, J. Baumgardner, W. Brewer, P. Gibson, and W. Remine. Using computer simulation to understand mutation accumulation dynamics and genetic load, in Y. Shi et al. (eds.), ICCS 2007, Part II, LNCS 4488, Springer-Verlag, Berlin, Heidelberg, pp. 386-392.
Joe G,
Hmm, maybe we could sequence the entire genome of humans and chimpanzees, compare them, then look for sign of natural selection in the genes that are different. I wonder if anyone has done that?
http://www.lcg.unam.mx/frontiers/files/frontiers/BustamanteMKPRF2005.pdf
Joe G,
ID may not be able to account for the origin of proteins, but etiobiology says they started as cofactors for RNA ribozymes.
William J. Murray,
I didn’t try to “prove a negative” William.
I just tried to explain why demonstrating that X is absent is so much harder than demonstrating that X is present.
Elizabeth,
Because something is harder doesn’t mean it cannot be done, nor does it relieve one of their obligation to support their assertions. If one cannot support a negative assertion – or a positive one, for that matter – they shouldn’t make it.
No, I quite agree that that does not mean it cannot be done. And I have agreed with you that several claims that you mention should not be made. For instance, the claim that science disproves ID (or even demonstrates that ID is false) is not a scientific claim.
Nor is the claim that current evolutionary theory satisfactorily accounts for every aspect of the origin of living things. Living things are very complicated!
ETA: and nor, of course, is the claim that an ID must have been involved. And I don’t just mean that in a woolly, “well we can’t be sure of anything” way. The big difference between ID and evolutionary theory is that ID claims that evolutionary theory is inadequate, therefore ID. At least in most of the papers I have read. Evolutionary theory does not claim that ID is inadequate or false (because, as I said, it’s unfalsifiable). It merely says that that we have a powerful theory that could explain the complexity and variety of life, but that there are some problems, not least the problem of how the first Darwinian-capable self-replicators arose.
I’ve got no trouble ascribing that to intelligent agency, but I would also hypothesize that said intelligent agency was itself a product of naturally occurring processes just as we are.
There would be no reason to propose otherwise.
William J Murray,
If the ID side cannot support the assertion, “an intelligent designer is required”, they shouldn’t make it.
Saying that unguided evolution is improbable, i.e., harder, doesn’t mean it can’t be done.
llanitedave,
“I would also hypothesize that said intelligent agency was itself a product of naturally occurring processes just as we are. There would be no reason to propose otherwise.”
There is no logical reason to require that presupposition.
That presupposition would a priori exclude the possibility of being able to test for that hypothesis based on objective evidence.
e.g. See Mark Herringshaw’s dissertation: “The Effects of Long-Distance Intercessory Prayer and Anti-Tobacco Communication on Teenager Intention to Smoke Cigarettes” “His ground-breaking research was the first known example of double-blind methods used to study prayer in medical science being applied in a social science discipline.”
Well the hypothesis was implicit in your carefully constructed example – that humans or something very like them drew these descriptions with the intelligent intent of providing a visual representation of something. One test would be to see if it could sensibly be mapped to what it seems to represent. Another test is to discover how the depictions came to be, learn their history. If someone drew them, find them and ask.
So was life Designed? Easy to test – just figure out its history according to the best evidence available. Which has been done pretty damn exhaustively, but apparently you don’t like the results. Another test – ask the Designer. But oops, we don’t mention this, ID doesn’t require it.
So let’s see…I’ve got it, let’s ASSUME design, disallow any investigation of the designer, the mechanisms of design or fabrication, or anything else salient. And THEN let’s insist that this position is the default UNLESS someone can PROVE us wrong! Consistent, predictive, thoroughly attested models, backed by millions of scientist-years of investigation, need not apply. That’s not how ID is done, don’t you know?
Yippee! Someone willing to state the obvious!
but I would also hypothesize that said intelligent agency was itself a product of naturally occurring processes just as we are.
How do you know we “are” the product of naturally occurring processes?
“but I would also hypothesize that said intelligent agency was itself a product of naturally occurring processes just as we are.”
Should be blockquoted above.
William J. Murray,
At one point you expressed strong interest in discussing Abel’s paper. There is now an entire thread dedicated to that. We haven’t sen you there, though. Are you no longer interested?
So, are you saying that the design inference is valid in the example I gave? Note: I’m not saying that ID was proven, only that it is a valid inference from the available information & evidence, and that such an inference leads to other areas of investigation?
I would suppose a design inference is valid in any circumstances where it can be falsified in principle.
But of course you understand that design inferences, like any other inferences, run the risk of false positives and false negatives. In general, the way to minimize such incorrect inferences is by learning as much about the subject as is possible. Mechanisms that can be demonstrated are important. Histories that can be reconstructed are important as well.
Often enough more than one interpretation of a given data set explains all that is known. Occam’s Razor isn’t the most reliable means ot picking one, but it’s often suggestive. But two VALID different interpretations necessarily imply a test to distinguish them. Properly operationalized, such a test can’t support both interpretations (it can support neither, and support is not proof).
So I agree with you so far. A design inference in your example seems reasonable. But if it cannot be tested and validated, it remains speculative. Validation requires a good deal of relevant evidence. If this is all we have, so that we can’t learn anything of the history, we are still guessing.
But does ID lead to other areas of investigation? I read that the Templeton Foundation has offered to fund any reasonable research to support a POSITIVE ID hypothesis. So far, not a single suggestion. I’m far from aware of all research, but the only ID research I’ve heard of has no positive ID hypotheses. It merely hypothesizes that some mechanism nobody ever posited, doesn’t work. And then proves it doesn’t work.
As a simple example, proving that 4+2 42, does NOT prove that 4+2=17.
I’ve read every post there with great interest. If I thought there was something significant or meaningful to the debate I could add, I would. There’s enough substance there to chew on for a while.
My interest was in getting someone to pause with the character attacks and hand-waving dismissals long enought to argue the merits to provide those of us not as educated in specific fields information to pursue and investigate. “Abel is an ID/Creationist Hack and You’re Too Uneducated to Realize This” isn’t an argument worthy of any serious consideration. Some of the posts in that thread are worthy of serious consideration and I appreciate those who took the time and effort to do so.
But I’m hardly qualified to argue that technical stuff one way or another; the best I could do is run down some of these arguments, investigate them, and ask more questions. I’m still in the process of trying to digest some of what is in there.
William J Murray,
I would say yes.