Saturday, December 08, 2012

Doctor Logic and Lydia McGrew on Likelilhoods, Design, and Probabilities

Doctor Logic: If an all powerful being were designing life, we don't expect descent, common descent, common composition or a gradual appearance of features and species. How many ways can a God create life in a universe? The number of ways a God can do this is vastly greater than the number of ways unguided evolution can do so. For example, gods don't even need to create life consistent with physical laws because they can create ghosts. There's no need for descent (birth) because God can make animals outright or create factories (no car has ever been born to another car). Even keeping the species the same and changing their natural histories and genomics gives a God vastly more options than evolution. I think theists would be tempted to say that there are infinitely more ways God could create life than ways that evolution could create life.


This is a simple problem in Bayesian reasoning. Finding ourselves in a world that is consistent with unguided evolution implies that the probability that we're designed is extremely close to zero.

In other words, if God exists, then there are a million ways in which God could create things, including Young Earth Creationism, etc. If atheism is true, then if intelligent life is going to emerge, it's got to emerge through naturalistic evolution. So, if the evidence is compatible with naturalistic evolution, then the evidence very strong supports naturalistic evolution, since this evidence is very likely given atheism and vanishingly unlikely given theism.

Lydia McGrew's paper on design and likelihoods might serve as a way for theists to respond here. Because God could do it a certain way doesn't mean that it would be reasonable for God to do so.


198 comments:

Crude said...

Man, so much wrong here. Where to begin.

For one thing, whether evolution is 'unguided' in the relevant sense is not established, nor can it be established, by science. Elliot Sober gives far from a non-exhaustive list of reasons why.

Second, multiple examples are red herrings or are simply confused. 'Gods don't need to create life consistent with physical laws because they can create ghosts.' is confused because granting the existence of ghosts would also grant the existence of laws governing their existence. It's not as if we consult a list of set-in-stone physical laws and go, aha, the physical laws were violated! We observe phenomena, and hypothesize physical laws that account for and include such.

Let me repeat that, since it seems as if so many people forget it: as far as science goes, 'physical law' is that which we describe to account for observations. When we encounter phenomena that is not subject to physical law as we understand it, the response is not always 'miracle!' but amending those laws.

Third, in principle, just about any means of creation you can imagine has an analogue in a hypothetical 'naturalistic' universe. You can have ghosts existence without God creating ghosts. You can have no origin of life, and life simply being past-eternal. You can even have factories, with the factories either law-spawned or merely existing. Hell, just look at the life theories prior to evolution: spontaneous generation, etc, comes damn close to 'factories making animals', and this wasn't some Christian or even theistic invention.

Common descent, common composition, even gradual appearances aren't some given on atheism/naturalism. Hell, 'intelligible origins' itself isn't even given on atheism/naturalism.

Once you clear away all the misconceptions, what you're left with is something like this: 'Naturalism is compatible with evolution!' And that is an extraordinarily low bar to jump.

B. Prokop said...

Good Grief! I guess Doctor Illogic would say that, given there were a million different routes I could have taken to get to Point B from Point A, then if I find myself at Point B, I must have gotten there by accident.

Gimme a break!

Karl Grant said...

Dr. Logic's talking about Bayes again? I suddenly have a headache.

ozero91 said...

"Good Grief! I guess Doctor Illogic would say that, given there were a million different routes I could have taken to get to Point B from Point A, then if I find myself at Point B, I must have gotten there by accident.

Gimme a break!"

Well, I think the question he is asking is credible. I think he is saying, of all the potential methods of creating life, why did God choose the a method that can be explained "naturalistically?" Though he needs to provide an argument as to why God, if he exists, HAD to use a method other then seting evolution in motion from a prime, genetic ancestor.

ozero91 said...

And by "seting" I mean "setting"

And maybe we should read the paper before addressing Logic's point.

Crude said...

Bob,

He's trying to make an argument along the lines of, 'Naturalism can only create one way. God can create in a billion ways, including naturalism's way. Therefore, if creation took place according to naturalism's way, you should suspect naturalism is true.'

I've pointed out problems with the reasoning presented. What's 'in principle' compatible with naturalism, particularly non-theism, is ridiculously greater than is being supposed here.

B. Prokop said...

I think he is saying, of all the potential methods of creating life, why did God choose the a method that can be explained "naturalistically?"

But under his "logic", he would be asking the identical question had God chosen some other way of creating life. ("Why this way instead of some other?" Because keep in mind that whatever alternate method was employed in such a thought experiment, that method would be "naturalistic" in that alternate reality.) That was the point of my rant. It's just another case of "Heads I win, tails you lose."

Karl Grant said...


You know, Dr. Logic tried to play a similar game about year and a half ago on this very site.

Jayman said...

How many ways can a God create life in a universe? The number of ways a God can do this is vastly greater than the number of ways unguided evolution can do so.

Shouldn't the analog to God be something like an atheistic universe? But if that's the case then you have a countless number of ways that life could arise naturally in such a universe.

I think theists would be tempted to say that there are infinitely more ways God could create life than ways that evolution could create life.

I am also tempted to say that there are infinitely more ways a godless universe could create life than ways evolution could create life.

If atheism is true, then if intelligent life is going to emerge, it's got to emerge through naturalistic evolution. So, if the evidence is compatible with naturalistic evolution, then the evidence very strong supports naturalistic evolution, since this evidence is very likely given atheism and vanishingly unlikely given theism.

What is the logical connection between atheism and evolution?

shiningwhiffle said...

@B. Prokop:

No, I think he's got a point. We could conceivably have found ourselves living in a world where the evidence for special creation was overwhelming: where, say, new things regularly come into being in a flash of divine light and with a booming voice saying, "I, the Lord, have now created this." But we didn't.

While I don't subscribe to falsificationism as the demarcation of science, there's a very real insight in it, since it's a special case of Bayesian reasoning. If an explanation is compatible with nearly any possible set of evidence we could find, finding that evidence doesn't strengthen our confidence in it much at all, while if another explanation is compatible only with a narrow range of possible evidence, finding that evidence strengthens our confidence in it quite a bit.

So I think Doctor Logic is right in that, if the evidence from science is really compatible with unguided evolution, that counts in favor of the notion that it really was unguided.

Karl Grant said...

Jayman,

But if that's the case then you have a countless number of ways that life could arise naturally in such a universe.

Good luck with that line argument. Me and about three other people tried that same track with Dr. Logic in the discussion I linked above. In fact, my exact quote was:

And since we are talking about ignorance and probability let's take a look at scientific theories into the naturalistic origin of life. We have the Urey–Miller experiment but now we think atmospheric conditions during Earth's early years were quite a bit different then those used for the experiment (even Dr. Miller admits his experiment proves nothing). We have the Panspermia theory, that life on Earth was seeded here via comet or asteroid but we are still left with the rich question of were the life on the comet/asteroid came from.

In fact, a quick Google search can provide you with some three or four dozen different competing theories on the the naturalistic origin of life and most of them have some pretty big gaps of knowledge in them. Now, according to the standards you are holding theism to, we should dismiss all naturalistic theories on the origin of life because of that ignorance.


He refused to get it then and he will refuse to get it now.

B. Prokop said...

"What is the logical connection between atheism and evolution?"

You've hit the nail on the head. There isn't any! A theist can believe in evolution (heck, the Pope does) and an atheist can believe in intelligent design (like a friend of mine who genuinely believes our universe is a computer simulation - I'm not kidding).

unkleE said...

I wonder what Dr Logic would say to this simple summary?

This is a simple problem in Bayesian reasoning. Finding ourselves in a universe that is consistent with cosmic constants that are extremely finely tuned to allow life implies that the probability that we're not designed is extremely close to zero.

Victor Reppert said...

The difficulties come in for conflict not with evolution per se, but with the thesis of unguided evolution. Remember the people who think you are an evolutionist if you don't accept the a literal six day creation. Well, you have atheistic evolutionists who think you are a creationist if you think God did anything in the process of speciation.

Sammy Onomato said...

Don't all orthodox Christians believe that God is the Creator? Even a Christian who completely accepts the scientific theory of evolution is still a creationist. I don't see why she should object to being called a creationist.

I think the real conflict arises when those who think God actively guided evolution try to present that belief as a scientific hypothesis.

Tony Hoffman said...

VR: "Well, you have atheistic evolutionists who think you are a creationist if you think God did anything in the process of speciation."

Well, that would make you a variety of creationist, wouldn't it? What else would you call special creation but a form of creationism? Evolutionary theory relies on unguided mutations, whereas the religious evolutionists, it seems, still believe in a kind of guiding -- I don't know what else to call this but a variety of creationism. I certainly can't see how one can accept modern evolutionary theory AND deny one of its most basic tenets.

B. Prokop said...
This comment has been removed by the author.
B. Prokop said...

All depends on how one defines one's terms (doesn't it always?).

If by "creationism" you mean "Did God create all things visible and invisible?", then count me amongst the creationists!

But that is not how the term is normally defined. In common usage (there is, after all, no official definition), a creationist is one who basically accepts Genesis as a literal account of how the universe came into being. By that definition, I am most emphatically not a creationist. No way, no how.

Tony writes, "I certainly can't see how one can accept modern evolutionary theory AND deny one of its most basic tenets." I agree. That is, if by "modern evolutionary theory" you mean "unguided".

But the very concept of "unguided" is a total departure from what we understand as the Scientific Method. No possible amount of empirical experimentation, examination of the fossil record, or mathematical modeling can ever answer the question of whether or not the evolutionary process was/is guided. That is purely within the realm of philosophy and/or theology.

Doctor Logic said...

Crude, I just gave a clear argument for how design can be established.

Ghosts: There will indeed be rules for a spirit world, but the rules would presumably be dualistic. Perhaps folk mental concepts like love and will might constrain reality, and free will would provide unpredictable variations. It doesn't matter. That's anything like the universe we live in. Physics in our universe is fundamentally non-mental.

No, there is no unguided universe analogue of factories in design. Yes, humans build factories in our universe, but that's because we designed them. Are you now going to tell me that if we discovered a primordial factory, ID advocates wouldn't jump on it as evidence for design?

You can have ghosts in physical universe if they're in a simulation, but not as a result of unguided evolution in a fundamentally non-mental universe.

Spontaneous generation, if it existed in a physical universe, would be like a fast-acting origin of life theory. It would be like RNA-world operating on the timescale of days. That's not design, but unguided abiogenesis. Surely, you can tell these apart.

Before evolutionary biology was put forward, design made sense, did it not? Where did these structures and mechanisms come from? If the universe was physical and life came into being unguided, there was no mechanism we could think of. Darwin created the first naturalistic synthesis of a solution. As far as I know, there are no naturalistic alternatives. Do you have more than I'm unaware of? Because Darwin's evolution gives descent, common descent, gradualism, etc.

B. Prokop said...

"I think the real conflict arises when those who think God actively guided evolution try to present that belief as a scientific hypothesis."

I also agree with this, with one caveat (to be dealt with at the end of this posting). The question of guided/unguided is not one that science can answer. Only philosophy/theology can deal with it. Real science should be silent on the issue.

Now for the caveat. I would change Sammy's comment to read as follows:

"I think the real conflict arises when either those who think God actively guided evolution try to present that belief as a scientific hypothesis, or those who believe in unguided evolution attempt the same."

Doctor Logic said...

Crude, let me put this more succinctly. Think of all the ways we might find strong and clear evidence of design. Primordial factories. Species with radically different genomes appearing literally overnight. Ghosts that violate our physical laws. There might be gaps and uncertainties in the design model, but we can imagine a wide array of vastly different sorts of experiences we could have.

Now think about the ways we would find strong and clear evidence for unguided evolution. The fossil record, DNA, common descent. There are some gaps in our knowledge, but we see what is expected under unguided evolution.

Do you really think there can be no evidence for design? That no world God would create would look obviously designed?

Doctor Logic said...

Bob,

"Good Grief! I guess Doctor Illogic would say that, given there were a million different routes I could have taken to get to Point B from Point A, then if I find myself at Point B, I must have gotten there by accident."

Well, Happy Sunday to you, too!

I don't see where "accident" comes into this. We're talking about standard inference to a theory from evidence.

Suppose I have two roulette wheels. One has 4 positions (numbered 1-4), the other has 4000 (numbered 1-4000). At random, I select a roulette wheel to play on. I play the selected wheel, and I report to you that my roulette wheel came up with the number 4. Which roulette wheel did I most likely play on?

And why?

Well, if I played this game with you 8000 times, then I would pick the 4-slot wheel 4000 times and the 4000-slot wheel 4000 times. In 4000 plays, I would get 4 a 1000 times on the 4-slot wheel. I would get a 4 on the 4000-slot wheel just once in 4000 plays. So, after 8000 games, we will expect 4 to come up 1001 times, 1000 of which will be from the 4-slot wheel. The odds that I played on the 4-slot wheel are 1000:1001. You infer with high confidence that I played the 4-slot wheel. It's not impossible that I played the 4000-slot wheel, but you would be irrational to infer that I had.

The bottom line: the theory which is more restrictive and consistent with the data wins against a theory which has less limited possibilities. And it's difficult to imagine a theory less limited than God.

Doctor Logic said...

Jayman,

It's not so much about atheism as about design.

This inference says nothing about the deistic God who didn't design life, for example.

Also, if we're living in a naturalistic universe, there might still be designed life, e.g., if we're living in a simulation.

Karl Grant said...

Suppose I have two roulette wheels. One has 4 positions (numbered 1-4), the other has 4000 (numbered 1-4000). At random, I select a roulette wheel to play on. I play the selected wheel, and I report to you that my roulette wheel came up with the number 4. Which roulette wheel did I most likely play on?

This again? You still haven't figured out that the decisions made by intelligent self-aware beings, like Bob deciding which route to take, are not in the same category as frigging dice throw or spin on the roulette wheel.

Doctor Logic said...

unkleE,

"Finding ourselves in a universe that is consistent with cosmic constants that are extremely finely tuned to allow life implies that the probability that we're not designed is extremely close to zero."

Well, I don't want to open this can of worms just yet. That's a separate inference. I'll just say that (1) our universe is not fine-tuned for angels, and (2) it's not remotely clear what the connection is between the laws and physical constants (fine-tuning may not even exist), and (3) we should be asking why our universe isn't a lot more fine-tuned for life (do angels get killed by tsunami in the spirit world?).

Doctor Logic said...

Karl,

"This again? You still haven't figured out that the decisions made by intelligent self-aware beings, like Bob deciding which route to take, are not in the same category as frigging dice throw or spin on the roulette wheel."

It doesn't matter. Probability is about ignorance. Suppose Bob could take any of a million routes, and he contemplates his choice according to numerous criteria, like the scenery, adventure, picking up his laundry, seeing unspecified friends, losing his wallet, whatever. As long as we are completely ignorant about what's going on in Bob's head, his choice is random to us, and the inference still applies.

Obviously, we can do much better than random when we know more about Bob. We might know that Bob looks out for his own safety, so he is less likely to take the dangerous routes. He looks out for his pocketbook, so he won't take the most expensive route, etc. These modifications to our probability estimate stem from Bob's goals, limitations and preferences.

The thing that makes this inference work is the fact that we're not saying anything about who the designer is. If we had a predictive model of the agent and his limitations/goals, we could narrow down what we would be likely to see. Yet, it is a strategy of the ID movement to avoid saying ANYTHING about the designer because they don't want to look like creationists.

ozero91 said...

"Our universe is not fine-tuned for angels."

How do you know this?

Doctor Logic said...

I just want to clarify that I don't endorse Bob's spoof of my inference.

Bob said
"I guess Doctor Illogic would say that, given there were a million different routes I could have taken to get to Point B from Point A, then if I find myself at Point B, I must have gotten there by accident."

A better analogy would go something like this. There are a million places Bob can go. He can go to Disneyworld or to the supermarket or to the beach. If we find Bob in the middle of the ocean, clinging to the life preserver of a recently sunken yacht, we infer that Bob arrived in this situation by accident (a sailing accident in this case). Of course, Bob could have deliberately decided that he wanted to simulate what happens to the survivors of boating accidents, or that he deliberately wanted to fake his own death. However, given that Bob could have gone bowling, to the shopping mall, or to countless other places with deliberation, we're pretty confident that he was in an accident.

You see, it's not just that there are a lot of things Bob could have done, but that specific accidents are quite predictive about where Bob will end up, whereas "whatever Bob feels like doing" is much less predictive.

Doctor Logic said...

ozero91,

"Our universe is not fine-tuned for angels."

How do you know this?


What I mean is that if I try to shoot or burn an angel, there's no requirement that the angel be harmed. Angels can be invulnerable to physics. If I beam an angel into the center of the Sun, it doesn't have to die. If I find a way to alter the physical constants of the universe, I can kill all life, but I don't necessarily kill all angels.

ozero91 said...

"Physics in our universe is fundamentally non-mental."

And doesn't this beg the question against, say, property dualism?

Karl Grant said...

Dr. Logic,

It doesn't matter. Probability is about ignorance.

Gee, I would never have figured that out. I mean, it's not like I said things like this the last time you tried this line of argument:

When all details concerning an event's probability are known the only options are 1 or 0, I believe I said that before. Inferences are the best guess we make when knowledge is incomplete. Knowledge varies from person to person; I know things you don't, you know things I don't

Or

then you should know that levels of knowledge and ignorance vary from person to person. Meaning somebody else is under no obligation to accept your personal view on the probability of theism versus atheism.

Oh wait, I did. The problem is that I already know from past experience that you are willing to ignore prior knowledge and evidence when it comes to making a probability distribution. That you prefer to live in the world of abstracts and theory as opposed to the world of hard evidence and concrete data.

Doctor Logic said...

ozero91,

I think property dualism is pretty silly, but it doesn't affect the argument I'm making. If property dualism were true, it wouldn't apply until life had evolved. Also, just because property dualism is possible doesn't mean substance dualism is impossible for God to create.

Karl Grant said...

Ozero,

How do you know this?

He doesn't, he's question begging.

Doctor Logic said...

Oh, my dear Karl,

You've just given the argument that, if God created the world, God knows he made an evolved world because he has more information than us. Congratulations!

ozero91 said...

"If property dualism were true, it wouldn't apply until life had evolved."

But if that were the case, wouldn't that be an example of non-material things (minds/mental properties) which are not subject to physical laws, appearing in our world without the need of a designer?

Doctor Logic said...

ozero91,

I guess I don't see the relevance.

The inference doesn't specifically relate to whether there are non-physical things, because the inference still works when there is a non-designer God.

Karl Grant said...

No Dr. Illogical,

I have given the argument that everybody has priors and that saying you think God is infinitely more complex than a naturalistic materialist interpretation of the evolutionary theory is you merely voicing your beliefs and we are under no obligation to accept them. I understand you didn't get this, I will strive to use less big words in the future.

Doctor Logic said...

Karl,

Some simple questions:

Is God capable of creating more kinds of life-bearing worlds than unguided evolution?

(Genesis seems to think so.)

Can God create life that doesn't fit an evolutionary tree? How many permutations of descent trees could God create?

Could God make animals that are radically different with no plausible evolutionary history?

Can God create life overnight?

Can God create angels invulnerable to physics in our universe?

Can God create monsters out of metal?

Can God create antelope that are powered by breeder fission reactors?

Karl Grant said...

Dr. Logic,

You are like a broken record, you know that? To quote JS Allen from the last time you dared bring this subject up over here,


adding more possibilities to what I might have made does not change your expectations one way or the other. If it does, you're bad at math.

You can't reduce someone's expectations that I made hamburgers by generating a bunch of imaginary recipes that I might make, just like you can't dilute away your bad math with an infinite number of gibberish arguments.


And

An infinite number of new possibilities doesn't infinitely dilute the odds, though you seem to desperately want to believe that.

Now how does that apply to what you just wrote? Besides the fact that what he told you went in one ear and out the other.

Sammy Onomato said...

B. Prokop said:

"I also agree with this, with one caveat (to be dealt with at the end of this posting). The question of guided/unguided is not one that science can answer. Only philosophy/theology can deal with it. Real science should be silent on the issue.

Now for the caveat. I would change Sammy's comment to read as follows:

"I think the real conflict arises when either those who think God actively guided evolution try to present that belief as a scientific hypothesis, or those who believe in unguided evolution attempt the same."


Would you object to a scientist saying that melting of an ice cube was an unguided process or that atomic decay was an unguided process?
Once it was thought that the planets were moved by supernatural agency, but we now know that such motion can be accounted for by an unguided process.

I don't see how acknowledging those processes as being unguided throw into question the belief that there is a god. Nor do I see how acknowledging that the processes involved in biological evolution are unguided throw into question the belief that there is a god.

ozero91 said...

Let's be charitable here guys, we're not gnus.

B. Prokop said...

"Well, Happy Sunday to you, too!"

I posted that on Saturday.

B. Prokop said...

Sammy,

Thanks for your answer. I actually do believe that all physical processes are "guided". This is part of what it means in the Creed where it states "Creator of all things visible and invisible" (emphasis added). Among the things invisible is all natural law.

So yes, I do regard the motions of the planets (or the melting of an ice cube)as guided. this is implied in St. Paul's phrasing "In [Christ] all things hold together" (Colossians 1:17). Please note that I am not arguing from Scripture here. I am making no argument whatsoever. I am merely stating my belief, in answer to your question.

ozero91 said...

Final causes

Sammy Onomato said...

B. Prokop,

Thanks for clarifying your religious belief.

But what I was trying to find out is whether or not you object to science describing the processes involved in planetary motion or the melting of an ice cube or radioactive decay as being unguided?

Seems to me that scientist's should be free set up their scientific models or explanations without regard to what any particular religious beliefs people may have. Scientist's describe all other physical processes as being unguided, so why should they be prohibited from describing the physical processes involved in biological evolution from being unguided?

Doctor Logic said...

Karl,

"To quote JS Allen from the last time you dared bring this subject up over here"

Oh, no! I dared to bring up a subject!

You can't reduce someone's expectations that I made hamburgers by generating a bunch of imaginary recipes that I might make, just like you can't dilute away your bad math with an infinite number of gibberish arguments.

Suppose Bob and Karl share a BBQ pit in their yard. Bob knows and loves 99 different BBQ recipes. Karl only knows how to make hamburgers. Suppose we know that either Bob or Karl (exclusively) cooked lunch at the pit today. The only evidence we have to determine the identity of the cook is our evidence that hamburgers were cooked today. Are you telling me that Bob's penchant for cooking a multitude of other things at the BBQ pit have no bearing on this inference?

If you are, then you are utterly irrational. Hypothetically, if we initially have no reason to believe that Bob cooks more frequently than Karl, then, if we inspected the BBQ pit on 198 occasions (99 occasions each cook), we would expect to find Bob cooking hamburgers once, and Karl cooking hamburgers 99 times. So, out of the 100 times we inspected the pit and found hamburgers, 99 of the times, Karl would be responsible. That's 99%, if you're paying attention.

If Bob knows and loves 999 recipes, then the probability that Karl cooked the hamburgers rises to 99.9%.

Are you getting this yet?

This is Probability 101.

Now, this all changes if you say that Bob LOVES hamburgers and cooks them 99.9% of the time. But if you do that, you're saying something predictive about the cook. You're saying that Bob can be expected to cook burgers 99.9% of the time.

If you refuse to say anything predictive about Bob's goals or propensities, then you have to spread the probability evenly across all the things Bob might cook. The increasing number of things Bob can/would cook makes it increasingly less likely that Bob was the cook.

Tony Hoffman said...

Bob Prokop: "So yes, I do regard the motions of the planets (or the melting of an ice cube)as guided. this is implied in St. Paul's phrasing "In [Christ] all things hold together" (Colossians 1:17). Please note that I am not arguing from Scripture here. I am making no argument whatsoever."

I can agree with that. But along with many of the other theist comments here, I am not sure what the objection is to Dr. Logic's argument.

If we cannot infer that evolution is more likely to be observed in a world where life arises from unguided forces than in a world created by an all powerful God, then what business does anyone have inferring that God has guided Evolution at all?

William said...

DL:
"
If you refuse to say anything predictive about Bob's goals or propensities, then you have to spread the probability evenly across all the things Bob might cook. The increasing number of things Bob can/would cook makes it increasingly less likely that Bob was the cook."

Something puzzles me here. This seems to turn the fine tuning of universal physical laws arguments on their ear, since it suggests that by increasing the number of potential alternative scenarios for something, we make random selection of the actual outcome more, not less likely.

I doubt you believe this about the universe, so can you explain how you mean it about terrestrial life forms?

Doctor Logic said...

Victor,

Lydia McGrew's paper helps me, not the theist.

McGrew is considering cases where we have some biological system which we have high confidence that evolution could not have produced. Such cases do not exist in fact.

It looks like the paper was written shortly after Behe's irreducible complexity was put forward. Perhaps this paper was intended to give philosophical support to Behe's design hypothesis. However, IC has since been refuted, so IC no longer stands up. There are unknowns in evolutionary biology, of course, but no cases where it is known that unguided evolution cannot achieve the structure.

In other words, McGrew is asking whether we can conclude design without saying anything about the designer from those special cases where unguided evolution does not provide an explanation. She says yes, and I agree with her. Paley's watch is a great example - there's no known unguided process that would create a watch mechanism (unguided evolution can create designers, but not directly wrist watches, as far as we know).

McGrew's paper actually supports my case because she points out the kinds of situations in which we could easily detect design using Bayesian methods. The generic case of design by a God would, all things being equal, provide us with countless such examples of design. Yet, it is these a priori predicted examples that we do not see.

Absence of evidence is evidence of absence, at least for generic designers.

Take a murder case as an analogy. Suppose we have lots of evidence the suspect is guilty (fingerprints, murder weapon, DNA). We can imagine scenarios in which we can show that our suspect is innocent without being able to name the actual murderer. However, if we lack an alibi/counter-evidence for our suspect, we would be irrational to conclude that our suspect isn't rather likely to be the murderer. Our suspect is millions of times more likely to be guilty than a generic member of the public. The generic member of the public is not predicted to leave fingerprints at the scene, for example. This is why people get convicted of murders based on evidence. It's not impossible that they were framed, but it's a lot less likely, all things being equal.

B. Prokop said...

Now you guys are just making me hungry!!!

Sammy, yes I would object to such an assertion - and strongly. "Science" has no business weighing in on the matter at all. That is precisely what is meant by your statement "Scientists should be free set up their scientific models or explanations without regard to what any particular religious beliefs people may have." Once a person starts claiming that an observed event is either guided or unguided, he is no longer remaining neutral vis a vis "religious beliefs people may have".

Surely that can be agreed upon?

Karl Grant said...

Dr. Logic,

One, William made a good point about what you just wrote.

Two, the fact that Bob knows how to cook other foods in addition to hamburgers does not preclude him from making hamburgers nor does it mean you ignore the possibility that he cooked the hamburgers nor does it mean you automatically assume Karl cooked them.

Three, I will repeat this, adding more possibilities to what Bob might have made does not change your expectations one way or the other.

Doctor Logic said...

William,

I don't understand your statement, but I'll say a few words about fine-tuning.

There are a couple of kinds of fine-tuning. The first kind has no bearing on life in the sense that it simply observes that physical constants have values that differ by many orders of magnitude. This is the thing that physicists want to clear up. It might turn out that a non-fine-tuned theory looks fine-tuned at low energy, and the problem might not be real.

The other kind of fine-tuning is the observation that if the physical constants were slightly different, either (1) no human life would exist, or (2) no life at all would exist.

It's not clear how much of the parameter space of physical constants might support life. If we could calculate physics in these hypothetical worlds, we would have been able to compute the low energy effects of string theory long ago. The fact that we can't predict physics with arbitrary constants means we can't justify (2).

As for (1), evolution predicts that we will be adapted for the universe in which we evolved. Theism doesn't, since, for example, angels don't need to be adapted to avoid freezing to death in the arctic.

Moreover, even if you suppose that a God wants to create life in a universe, fine-tuning laws of physics for evolution doesn't make a better choice than fine-tuning laws of physics and creating life in some other fashion.

B. Prokop said...

"Then what business does anyone have inferring that God has guided Evolution at all?"

Scientifically, none at all. What's sauce for the goose, after all. The inference that natural processes are guided is a theological statement - not a scientific one.

I see zero conflict of interest here. Science teaches me how my body works. It says nothing about why I have one.

Doctor Logic said...

Karl,

You are simply wrong. Go and talk to any statistics professor you want.

I DID account for the possibility that Bob cooked the hamburgers. That's why I concluded a 99% probability Karl was the cook and not a 100% probability. See?

Doctor Logic said...

Bob,

Read McGrew's paper. Don't you agree there can be scientific evidence for design versus unguided evolution?

If there can be scientific evidence for design, there can be scientific evidence for its absence.

B. Prokop said...

No, I emphatically reject the notion that there can be scientific evidence of design. All we can do by empirical study is ascertain what level of complexity exists. What we cannot do is determine what level can be expected given purely naturalistic assumptions.

That said, I personally find the Argument From Design to be persuasive. It is one of the many, many reasons I am comfortable being a Catholic Christian instead of an atheist. But I do not regard the argument to be "science".

Doctor Logic said...

Bob,

You sound like you're just blindly saying you prefer P to ~P, but the evidence doesn't count either way.

Would you say that God can't create any more kinds of life than unguided evolution?

If so, you're making a Bayesian commitment.

Karl Grant said...

Dr. Logic,

And what would you say if I told you I showed your comments last year to two different mathematics professors and both said you were a confused individual?

ozero91 said...

Crude, where is your first link supposed to go?

ozero91 said...

Ah, nevermind, I saw the link to his paper.

Also, Methodological Naturalism v.s. Metaphysical Naturalism, etc.

B. Prokop said...

"Would you say that God can't create any more kinds of life than unguided evolution?"

I wouldn't answer that question at all. Such semantically null propositions remind me of St. Paul's advice:

"Avoid the godless chatter and contradictions of what is falsely called knowledge." (1 Timothy 6:20)

"Avoid disputing about words, which does no good" (2 Timothy 2:14)

"Avoid foolish controversies ... These things are useless and a waste of time." (Titus 3:9)

Gimme something I can work with!

Karl Grant said...

And to your BBQ example Dr. Logic, the odds that Karl made the hamburgers instead of Bob is closer to fifty percent. The scenario is this:

1) Hamburgers are cooked in the BBQ Pit.

2) Karl and Bob both use this BBQ Pit.

3) Both Karl and Bob know how to make hamburgers.

The fact that Bob knows how to cook more food then Karl is complete and utter red herring. The fact that Bob knows how make baby-back ribs and Karl does not doesn't matter. What matters is we know somebody cooked hamburgers and we know both potential cooks know how to do that. Go read an elementary math text book.

B. Prokop said...

You guys are all missing the obvious! The easiest way to figure out who cooked the hamburgers is to taste one. If it's the best damn hamburger you've ever eaten, then I cooked 'em. Otherwise, Karl's your man!

Selah.

Jayman said...

Doctor Logic, I am trying to attack your argument at what I take to be its most fundamental level.

You seem to be saying the following:

(1) The probability of evolution on atheism is high (1 in 4?)
(2) The probability of evolution on theism is low (1 in 4000?)

I see no reason to accept your numbers. I would counter that:

(3) The probability of evolution on atheism is unknown.
(4) The probability of evolution on theism is unknown.

B. Prokop said...

I have always found these discussions about probabilities to be utterly pointless. I agree with Jayman that people just seem to plug in whatever numbers they feel bolster their going-in premises.

Just as in all the endless discussions I've seen on this website about the so-called probability of the Resurrection. As I see it, it either happened or it didn't, making the probability of its occurrence either one or zero.

Doctor Logic said...

Karl,

And what would you say if I told you I showed your comments last year to two different mathematics professors and both said you were a confused individual?

Then I would think you a liar.

Karl Grant said...

Dr. Illogical,

Yeah, that was the response I expected.

Doctor Logic said...

Jayman,

(1) The probability of evolution on atheism is high (1 in 4?)
(2) The probability of evolution on theism is low (1 in 4000?)

I see no reason to accept your numbers. I would counter that:

(3) The probability of evolution on atheism is unknown.
(4) The probability of evolution on theism is unknown.


First, this isn't about atheism because it's possible we're accidental, even if a god exists. Maybe god doesn't care what happens. This is about design versus non-design.

Second, knowing the probability is unimportant. What we care about are the relative probabilities.
http://en.wikipedia.org/wiki/Bayes'_theorem

All things being equal, this can be estimated by comparing the number of possible outcomes consistent with each theory.

In the BBQ example, suppose I don't know how many recipes are known to Bob & Karl. In that case, I don't know the probability of Bob or Karl cooking any particular meal. However, suppose that, for every recipe Karl knows, I can point to 99 that Bob knows. In that case, as long as the evidence is consistent with Karl cooking the food, the inference is exactly the same. The probability that Karl is the cook is 50/50 a priori, but 99/100 after we find it's one of the relatively few meals Karl can cook.

Karl Grant said...

Of course, JS Allen, who had a math degree and twenty years work experience with probability distributions said the same thing about you in the last debate and that didn't seem to phase you.

Jayman said...

Doctor Logic:

This is about design versus non-design.

Then I'll re-phrase my objection:

(3) The probability of evolution on non-design is unknown.
(4) The probability of evolution on design is unknown.

All things being equal, this can be estimated by comparing the number of possible outcomes consistent with each theory.

As far as I can tell, the number of "possible outcomes" for either scenario is infinite.

Crude said...

Crude, I just gave a clear argument for how design can be established.

No, you didn't. Let's take a look at some of your followup explanation to see where the problems are.

There will indeed be rules for a spirit world, but the rules would presumably be dualistic.

First, that really depends on what you mean by ghosts. The ghosts of ghostbusters? The ghosts of popular imagination? No, that actually isn't 'dualism', either cartesian or hylemorphic. It's beings made out of some odd kind of matter.

Second, neither this nor the assumption that full-blown cartesian dualism would be in play, matters for your claim. Let's have a look at it.

Perhaps folk mental concepts like love and will might constrain reality, and free will would provide unpredictable variations. It doesn't matter. That's anything like the universe we live in. Physics in our universe is fundamentally non-mental.

I assume by 'anything' you mean 'nothing'. First, some physicists disagree about our universe being 'fundamentally non-mental' - that's not established, and there have been some pretty weighty, recent dissenters. (Wigner, etc.)

Second, here's why it doesn't matter: because it's entirely consistent with atheism that such universes exist. Remember, your standard here is 'a universe designed by God' and 'a universe not designed by God'. For 'not designed', you're dealing with a metaphysic entirely at home with brute facts.

Hell, I notice you concede right away that *design* is compatible with atheism: you count simulated universes as under the 'atheist' standard.

Are you now going to tell me that if we discovered a primordial factory, ID advocates wouldn't jump on it as evidence for design?

ID advocates jump on all kinds of things, now, as evidence for design. That means nothing to you. In fact, you have a ready reply: 'It doesn't mean God designed it.' You already gave it without me having to even press much.

Spontaneous generation, if it existed in a physical universe, would be like a fast-acting origin of life theory. It would be like RNA-world operating on the timescale of days. That's not design, but unguided abiogenesis. Surely, you can tell these apart.

No, it's not 'unguided abiogenesis' in principle - abiogenesis can in principle be 'guided' in the relevant sense. Are you sure you recall what spontaneous generation was?

Crude said...

Before evolutionary biology was put forward, design made sense, did it not?

It makes sense now. What's more, you cannot possibly be conflating evolutionary biology with the origin of life, can you? Every evolution advocate around goes through pains to distance those two topics.

As far as I know, there are no naturalistic alternatives. Do you have more than I'm unaware of?

There are plenty of 'naturalistic alternatives'. Hell, you yourself gave some. I gave plenty.

Think of all the ways we might find strong and clear evidence of design. Primordial factories. Species with radically different genomes appearing literally overnight. Ghosts that violate our physical laws. There might be gaps and uncertainties in the design model, but we can imagine a wide array of vastly different sorts of experiences we could have.

I've pointed out that all of these are entirely consistent with an atheistic universe in principle, while you yourself went further and said that even if you have evidence for design, that design could be by a being other than God.

You fail to understand - or at least fail to acknowledge - that your argument is about what odds are for such and such, given two *extraordinarily broad* metaphysical views. Atheism is compatible with everything from factories creating life (aka, spontaneous generation), to dualism, to 'violations of physical law' (aka, new scientific discoveries.)

Evolution is not unguided, nor can it be established as unguided, if by that you mean 'God neither guides nor knows the outcomes of it'. Again, I reference Sober.

Finally, I have to ask: are you arguing that 'God of the gaps' is legitimate? All of your examples seem to turn on you saying that phenomena X could take place with no naturalistic explanation for X, therefore God. While I've pointed out that that's false in terms of the 'no naturalistic explanation part', are you really saying 'we lack a naturalistic explanation' suffices as evidence for God's existence?

Crude said...

Do you really think there can be no evidence for design?

Sure I can. I simply recognize that 'no naturalistic explanation for X is possible' or worse, 'no naturalistic explanation for X is possible, even in principle!' are complete and utter red herrings. I know that 'naturalism' is so extraordinarily broad that it's in principle compatibility is considerable. Especially when you yourself are saying (as, really, you have to) that 'evidence for design' is compatible straightaway with 'atheism', and your argument was explicitly about God.

I know that what 'looks designed' is going to rest on metaphysical priors, assumptions about the natural world, and even our own experience as designers. A plastic plant looks damn designed to me for obvious reasons. For some tribal person? It's just some weird-ass plant.

That no world God would create would look obviously designed?

Except what 'looks obviously designed' is itself subject to furious disputes, even now. Does a cell look obviously designed? ID proponents say yes, and they find the attempted explanations of various aspects of the cell laughable. ID critics often say no.

How about your genetic code? Again, plenty of people say yes, it's obviously designed - hell, the fact that biology makes use of coded language should itself shell-shock anti-design proponents. But anti-design proponents complain that they think the code is inefficient, design proponents dispute that and point out that not all design is good design. And on and on and on and on it goes. Before we even get to 'designed doesn't mean designed by God, which was your original standard'.

Tony Hoffman said...

Jayman, thank you for actually engaging with the argument. I find this topic interesting, and it would have been a shame if no one was able to come up at least valid-seeming objections.

Doctor Logic said...

Karl,

I apologize. Jumping to "liar" was wrong, and I feel bad about that.

I don't think that what you've said is the whole truth, even if you believe it is. You're confused, and, perhaps, if you did try to present this issue to math professors, you failed to do so accurately. Or maybe these professors don't know what they're talking about.

Doctor Logic said...

Jayman,

As far as I can tell, the number of "possible outcomes" for either scenario is infinite.

This doesn't affect my argument. Just imagine that no two hamburgers are exactly alike. That is, consider "hamburger" or "ribs" each refer to N recipes where N tends to infinity. You'll see that using continuous distributions doesn't affect the result because N drops out.

Karl Grant said...

Dr Logic,

You're confused, and, perhaps, if you did try to present this issue to math professors, you failed to do so accurately. Or maybe these professors don't know what they're talking about.

I showed them the discussion online and had them read your words at the source. Also, the majority of people that have commented on both this discussion and the previous one have not taken you side. If I am confused, I am certainly not the only one.

Doctor Logic said...

Crude,

I've gone back to my original comment, and it seems that you have misunderstood me quite thoroughly.

This is about design versus non-design, not naturalism versus dualism, or theism versus atheism.

The issue is whether life was foreseen and is the result of a plan. My point is that a designer has far more options than an unguided system without forethought.

I disagree with you about the nature of Ghostbuster-style ghosts, but it just doesn't seem particularly relevant. If a designer wants a universe of thinking beings, Ghostbuster-style ghosts are an option. They're not an option under the standard theory of biology, and they don't appear to exist in reality (signal well below the noise).

If my argument works, it does not imply no god exists. It implies a designer god was not involved in creating life.

ID advocates jump on all kinds of things, now, as evidence for design. That means nothing to you.

Right. It means nothing. They have to do better than "Hey, ignore all the positive evidence for unguided evolution, and look at this here gap!!!"

The problem with ID advocates is that they think the evidence for evolution to date has no bearing on design, but a gap in the evidence implies design. They can't have it both ways.

Crude said...

DL,

I've gone back to my original comment, and it seems that you have misunderstood me quite thoroughly.

This is about design versus non-design, not naturalism versus dualism, or theism versus atheism.


That doesn't affect most of my reply, but really - you mention 'God' or 'gods' ten times in the OP quoted statement. If you're telling me that you didn't mean to talk about God/gods, but design, I think you should have been clearer.

Especially when you say this: If my argument works, it does not imply no god exists. It implies a designer god was not involved in creating life.

There you are, back to God/gods again.

The issue is whether life was foreseen and is the result of a plan. My point is that a designer has far more options than an unguided system without forethought.

And as I've said repeatedly: science does not establish, nor can it establish, that evolution is unguided and without forethought, in the relevant senses. That's not just my view - that's the view of the NCSE, Eugenie Scott, and more. (Not that their dissent would matter for my purposes, but the NABT controversy does help illustrate the point.)

If a designer wants a universe of thinking beings, Ghostbuster-style ghosts are an option. They're not an option under the standard theory of biology, and they don't appear to exist in reality (signal well below the noise).

The 'standard theory of biology' is based on known phenomena. We're talking about in principle possibilities given an undesigned universe and a designed universe.

I'm pointing out - take an alternate universe. Ghosts exist - strange odd beings that live in another plane of existence, etc. What, precisely, requires them to be designed? The answer is, nothing, in principle. Certain laws may simply exist. Certain natures may simply be.

That's what you're not appreciating: 'non-designed' is ridiculously broad in scope, in principle.

Crude said...

Right. It means nothing. They have to do better than "Hey, ignore all the positive evidence for unguided evolution, and look at this here gap!!!"

First off: you endorse "gap" thinking. That's a major thrust of your argument here. Even though I say it fails on the level you're using it, you can't complain at once that ID proponents rely on gaps, yet say that if the universe is designed we should find gaps.

Second, as stated - 'unguided' evolution isn't supported scientifically in the sense you mean.

Third, ID proponents do far, far more than argue about gaps. They call on the demonstrable existence (at least demonstrable if you're not a materialist a la Rosenberg) of intelligent design in our universe, and make arguments of analogy. They examine the data from Lenski, etc, and point out the problems - they don't merely say 'this isn't explained yet', they argue why they think theory X cannot suffice to explain the phenomena.

The problem with ID advocates is that they think the evidence for evolution to date has no bearing on design, but a gap in the evidence implies design. They can't have it both ways.

Not true, but more than that, you're confused on this subject. First off, many ID proponents DO think that evolution has a bearing on design - what they deny is the state of the evidence for it in the relevant senses.

Second, as I've pointed out, science is utterly quiet on 'design' in the relevant sense you're discussing here.

Third, 'evolution' is too broad, since numerous ID proponents think their views are compatible with evolution. What they reject is a particular theory of evolution.

And even with all this in mind, the fact is you're still advocating God-of-the-gaps thinking (even demanding it) even while decrying it, and you still haven't faced up to the fact that the spread of what's in principle possible in a non-designed universe is tremendously broad - it's not limited to what you wrongly call 'unguided evolution'.

Doctor Logic said...

Crude,

http://en.wikipedia.org/wiki/Spontaneous_generation

I assume you brought this up because you think that it represents a general law that life just comes into existence without design. Perhaps you offer this as a non-theistic solution or a naturalistic solution.

Again, why is this relevant?

We're comparing the theory of evolution with a generic design theory. We're not comparing an unspecified naturalistic or atheistic theory against an unspecified designer/theistic theory. If we were, perhaps the odds would be even, though I doubt that.

No, we're comparing a specific theory to alternatives in light of the evidence. We have the neo-Darwinian synthesis, which makes rather definite predictions, all of which have been verified.

ID advocates don't want to put forward a specific theory of design which would require talking about a designer. Since they say nothing about a designer, we have to integrate over all possible designers of life.

Generically, then, design is ruled out. It could only make a comeback if you cook up an ID theory that is as predictive of what we see as NDE. Good luck with that.

If it makes you feel any better, NDE also competes with other naturalistic theories, designed and otherwise. Spontaneous generation is(was) a competitor, and it, too, has been crushed by NDE as an explanation for speciation.

Here's one more way of looking at this. In Bayes' theorem will appear the theory T, where T refers to unguided NDE. In the denominator, we'll find ~T. ~T includes design and non-design explanations that are not NDE. If you think there are non-design explanations for anything we could possibly see, I'm just as happy, because they are ruled out in exactly the same way.

ozero91 said...

"No, we're comparing a specific theory to alternatives in light of the evidence. We have the neo-Darwinian synthesis, which makes rather definite predictions, all of which have been verified."

Predictions like, "If birds descended from therapods, then we would expect to see fossils of transition species in this time period." are perfectly compatible with a sort of "divine plan."

What sort of predictions are you refering to? "If evolution is not part of a "divine plan" then... what exactly?"

Crude said...

I assume you brought this up because you think that it represents a general law that life just comes into existence without design. Perhaps you offer this as a non-theistic solution or a naturalistic solution.

Again, why is this relevant?


Because it helps to illustrate that the space of possibilities for life in a non-designed universe are tremendous, whereas your argument requires that they be few. I'm pointing out that this is flawed reasoning.

Further, do you dispute Victor's summary of your argument in the OP?

No, we're comparing a specific theory to alternatives in light of the evidence. We have the neo-Darwinian synthesis, which makes rather definite predictions, all of which have been verified.

Er, the big giveaway there is 'neo-Darwinian synthesis'. Do you know what it's neo-Darwinian rather than 'Darwinian'? Because evolution as Darwin conceived it was riddled with flaws. As for the health of the neo-Darwinian synthesis, that's disputed even by ID Critics.

But, that's besides the point.

We're comparing the theory of evolution with a generic design theory. We're not comparing an unspecified naturalistic or atheistic theory against an unspecified designer/theistic theory. If we were, perhaps the odds would be even, though I doubt that.

The theory of evolution, insofar as it's a scientific theory, is utterly silent on whether or not evolution is a guided process with intended outcomes. It does make some general claims about certain known correlations, but not with regards to 'Was this evolutionary result intended or unintended by a designer?'

If you contest this, my request is simple: show me the peer-reviewed, published research article testing the hypothesis 'Are evolutionary outcomes following the intentions of a designer?' I'd love to see that paper.

Jayman said...

Doctor Logic, would you care to post the relevant equations (as you see it)?

Doctor Logic said...

Crude,

Finally, I have to ask: are you arguing that 'God of the gaps' is legitimate? All of your examples seem to turn on you saying that phenomena X could take place with no naturalistic explanation for X, therefore God. While I've pointed out that that's false in terms of the 'no naturalistic explanation part', are you really saying 'we lack a naturalistic explanation' suffices as evidence for God's existence?

No. A gaps argument asks for as-yet-undiscovered details of an already confirmed theory.

NDE is unguided. The ID god-of-the-gaps argument says, "hey, I know we've only just met, and all the evidence in the tree of life points to unguided NDE, but this thing that hasn't been explained in detail... it's design. Never mind that design doesn't need NDE."

In a murder investigation analogy, imagine we have lots of evidence that we have the right suspect. Fingerprints, DNA, murder weapon, eyewitness reports. A gaps argument for his innocence would be saying that we haven't discovered how the suspect traveled from his home to the murder scene, therefore, he's not guilty.

Of course, this would be a silly defense. Exculpatory evidence would instead take the form that it is impossible for the suspect to have committed the crime, not merely that we don't know some of the facts yet. And this evidence has to have the comparable Bayesian inferential weight to the condemning evidences.

This is what McGrew is arguing for, if I understand her paper.

Crude said...

No. A gaps argument asks for as-yet-undiscovered details of an already confirmed theory.

No, 'asking for details' is pretty much the exact opposite of a gaps argument. As usual, poor cite, but the wikipedia puts the God of the gaps argument essentially as, 'X is not understood/explained by science. Therefore, God.'

"An already confirmed theory." has little to do with it.

The ID god-of-the-gaps argument says, "hey, I know we've only just met, and all the evidence in the tree of life points to unguided NDE, but this thing that hasn't been explained in detail... it's design. Never mind that design doesn't need NDE."

Not as ID proponents put it. Their view is:

Phenomena X exists. We have demonstrable analogous evidence that process A (intelligent design) is capable of accounting for phenomena X. The evidence indicates that process B (the proposed mechanisms of this particular evolutionary theory) is not capable of accounting for phenomena X. We know of no other alternatives at the moment. Therefore, we infer process A.

You may find flaws with any aspect of their reasoning, but that's far closer to how they present their argument.

Doctor Logic said...

ozero91,

Predictions like, "If birds descended from therapods, then we would expect to see fossils of transition species in this time period." are perfectly compatible with a sort of "divine plan."

Compatibility is irrelevant. Probability is everything.

Every murder investigation has been compatible with framing of the innocent suspect by an infinitely devious Illuminati.

The point is that design doesn't need birds to evolve from therapods. It doesn't need therapods at all, and doesn't even need a fossil record because life could be deployed overnight.

We can fine-tune our design theory, and suppose the designer wanted his design to look like evolution (his "divine plan"), but if we do that without evidence, we become conspiracy theorists. It's like saying that every inference from a criminal trial is just an example of the Illuminati's "devious plan".

Doctor Logic said...

Jayman,

Doctor Logic, would you care to post the relevant equations (as you see it)?

Sure.

Bayes' theorem refers to three kinds of quantities.

(1) P(T): Prior probability of a theory, T, being true.

(2) P(O|T): the likelihood of making observation O if theory T is true.

(3) P(T|O), the conditional (or updated) probability of theory T being true, given observation O.

Generally, Bayes' theorem looks like this:

P(T|O) = P(T) P(O|T) / [P(T) P(O|T) + P(~T) P(O|~T)]

Suppose Tk is the theory that Karl cooked at the BBQ pit. Tb is the theory that Bob cooked at the BBQ pit. For our purposes, T = Tk, and ~T = Tb.

Initially, we think it's 50/50 that Bob or Karl was the cook, so P(Tk) = P(Tb) = 0.5.

These prior probabilities drop out of the equation.

Recall that our evidence is that hamburgers were cooked today at the BBQ pit. Recall also that Bob knows and likes 99 recipes, while Karl knows only burgers.

This means P(burgers|Tk) = 1
and P(burgers|Tb) = 1/99.

Plug this into Bayes' theorem and we find our prior probability of 50% that Karl was the cook jumps to

1/ [1 + 1/99]

= 1 / [99/99 + 1/99]

= 1/ [100/ 99]

= 99%.

Now, you could suppose that I have assumed that there is only one kind of burger, i.e., that all burgers are indistinguishable. However, we know that bread, meat and other ingredients might be combined in slightly different qualities and quantities. We might suppose that these quantities and qualities are continuous and infinitely variable.

To simplify the math, let's suppose that all recipes can use USDA grade A beef or grade A+ beef. In that case, Karl actually knows two recipes, not one: grade A hamburgers and grade A+ burgers. Likewise, Bob actually knows 198 recipes because he can use grade A or A+ beef in all of his recipes.

This means P(burgers|Tb) = 1/99x2
and
P(burgers|Tk) = 1/2.

The factor of 2 falls out of the calculation.

The factor will always fall out of the equation as long as the number of variations of each recipe is comparable, e.g., Bob is assumed to know grade A and grade A+ ribs recipes.

We can imagine adding more variations. Say 1/2 lb beef burgers and 1/4 lb beef burgers. Plain buns and sesame seed buns. As long as we can find comparable variations of all the recipes, the factor of N falls out of the equation, and it doesn't matter if the number of variations tends toward infinity.

Crude said...

DL,

Bayes' theorem refers to three kinds of quantities.

First, I think Jayman was asking for the probabilities of your argument regarding design. Not burgers.

Second, your argument about burgers seems to go back to saying something along the following lines:

"If life isn't designed, then in principle the only way these various species we see could be explained is by evolution. If life is designed, then in principle there's a variety of ways these species could be explained."

But that lands you right back with all the criticisms I pointed out of that position since comment 1.

To put it in a way appropriate for your example: if Karl actually knows 99+ recipes, your argument falls apart. If your argument is that the only possible way various species could exist in a non-designed universe is by neo-Darwinian evolution (1 recipe), then I've pointed out a legion of objections to that claim.

Doctor Logic said...

Jayman,

An objection to the above would be to suggest that there are infinitely many more burger variations than ribs variations. I don't think this is true, assuming there are no zero-mass, zero-volume ingredients. Basically, a burger weighing 1 lb may have more ingredients than ribs weighing 1 lb, but ribs have a comparable amount of tissue variation which cancels out the factor of N as N goes to infinity. This means burgers might end up with a likelihood weight of 2 (or 10) relative to ribs, but not an infinite weight.

We only have to consider relative volumes of spaces, not absolute volumes.

Doctor Logic said...

Crude,

To put it in a way appropriate for your example: if Karl actually knows 99+ recipes, your argument falls apart.

It's true that if Karl knows 99+ recipes, the inference breaks down. However, I think you've missed the analogy.

You incorrectly think Tk corresponds to "The theory that the universe was not designed."

I've already explained that Tk corresponds to "The theory of unguided NDE."

If your argument is that the only possible way various species could exist in a non-designed universe is by neo-Darwinian evolution (1 recipe), then I've pointed out a legion of objections to that claim.

Certainly, there might be non-design scenarios (spontaneous generation?) in which there are lots of species, but spontaneous generation does not preferentially predict common descent or the observed fossil record. Of course, if you've found an alternative non-design explanation for observed natural history, I (and the world) would love to hear about it. ;)

Tb in this analogy is not just design, but non-NDE non-design.

Jayman said...

Doctor Logic, Crude is correct that I was looking for the equation concerning design. Nonetheless, here is how I see equation working out.

T = theory that life is designed
O = evolution
P(T) = 0.5
P(O|T) = ?
P(T|O) = ?

This also means that we don't know the P(~T|O) where ~T is the theory that life is undesigned.

But it appears that you are comparing the theory that life is designed with the theory that life came about by undesigned neo-Darwinian evolution. It seems to me that this is an apple-to-orange comparison.

Crude said...

It's true that if Karl knows 99+ recipes, the inference breaks down. However, I think you've missed the analogy.

You incorrectly think Tk corresponds to "The theory that the universe was not designed."

I've already explained that Tk corresponds to "The theory of unguided NDE."


It's an invalid comparison, especially if Tb is 'The theory that the universe was designed.'

Certainly, there might be non-design scenarios (spontaneous generation?) in which there are lots of species, but spontaneous generation does not preferentially predict common descent or the observed fossil record.

Look, here's your problem.

If your comparison is between 'The general theory that the universe is designed' and 'The general theory that the universe is not designed', and you believe that the latter is 'NDE', while the former is some huge variety of options, you're wrong for the reasons I've pointed out already.

The only way specifying 'NDE' matters is if 'the general theory that the universe is not designed' cashes out to 'NDE, period'. If Karl actually knows 99 recipes, and among Bob's recipes is a recipe for burgers, then 'Someone is cooking burgers' isn't of much help in determining whether Karl or Bob is cooking.

Of course, if you've found an alternative non-design explanation for observed natural history, I (and the world) would love to hear about it. ;)

What's relevant here is what we should expect to find given the spread of in-principle possibilities for a designed versus undesigned universe, as detached from metaphysics as we can manage. 'Evolution and common descent' are common to BOTH the designed and non-designed spread of possibilities. 'Something other than evolution and common descent' are likewise common to both views.

You even seem to cop to this re: spontaneous generation, but you seem to think that doesn't matter so long as current observations support NDE. I'm pointing out you're incorrect. You're further incorrect with regards to 'unguided' as far as God is concerned - again, I reference Eugenie Scott and the NABT case for, at the very least, some atheist biologist support on this front.

Tb in this analogy is not just design, but non-NDE non-design.

I think you mean non-NDE design. But even that would be flawed, since 'NDE' is subsumed under the design possibilities as well - at least insofar as scientific NDE is concerned.

William said...

DL:

The problem with Bayes calculations is that they may ignore multiple pathways of causality. In your case,I see now, you are arguing that God is much less likely to choose evolution to create life than a non-agent cause would have to result in life by chance.

This deserves a causal map for better analysis (see J. Pearl et al, on causality).

Here's a possible diagram, ignoring the unlikely causalities,such as life causing the universe backwards in time:

http://lynxview.com/img/etc/lifefrom.jpg

As you can see,our probabilities are with C = divine creation, PPU = universe prior to life, E = evolution, LHN = life here and now.

The probability vector requires we calculate

P( C | LHN )


...and I thnk you are saying that
P(E|C) is low compared to the influnce of P( E | (PPE & ~C ).

I think that there are so many unmeasured influences here that it's all quite indeterminate. All of the factors are greatly dependent on the prior value P( C ) which is exactly what is being argued over...

Doctor Logic said...

Jayman,

Crude says:

First, I think Jayman was asking for the probabilities of your argument regarding design. Not burgers.

Fair enough. In considering alternative designs, one possibility is that of separate creation of different genera or species. A designer could create all the species on Earth with separate genomes. Or not. What is the number of permutations of genomes a designer could create relative to what NDE could create?

Well, at the very least, for every species, a designer could substitute a radically different genome (or not). This would give us a factor of at least 2 to the N variations of species where N is the number of species. This is an absurdly conservative estimate. NDE can't do this substitution at all.

A designer could preserve the number of species, guaranteeing that any buffalo killed on Earth is replaced by a new buffalo somewhere else on Earth. This could be true of some species and not of others. Throw in another factor of 2^N. NDE cannot do this.

Factories. For every species, a designer could implement (or not) a factory instead of a reproductive scheme. For example, turtles could lack reproductive organs, and instead we find designed/constructed/non-evolved factories that make turtles. For every observed species, there is a factor of 2 in number of variations (for factory or reproduction) yielding a factor of 2 to the N is the number of species. NDE cannot do this.

We can imagine a designer creating vastly more or fewer species than NDE can achieve. We can imagine a designer implementing these on timescales much faster than NDE can achieve (e.g., instantaneously, as Genesis says).

A designer can create species that are made of non-naturally occurring materials, e.g., pure silicon wafers or pure, annealed materials that are incompatible with NDE.

Simply put, NDE places a lot of restrictions on what kinds of natural histories can exist. A designer doesn't have these limitations. I think really smart designers can think up even more peculiar arrangements that are out of reach of NDE.

Crude said...

Simply put, NDE places a lot of restrictions on what kinds of natural histories can exist. A designer doesn't have these limitations. I think really smart designers can think up even more peculiar arrangements that are out of reach of NDE.

And you're right back to comparing what's possible given NDE given what's possible given 'any kind of design in principle' - despite that being an invalid comparison. 'Non-design' scenarios could in principle cover pretty much everything on your list - when you're talking about hypothetical universes, you're dealing with an incredibly elastic concept.

Worse, your examples get into tautologies. For instance:

For example, turtles could lack reproductive organs, and instead we find designed/constructed/non-evolved factories that make turtles.

Yes, if you build in 'designed' into the factory idea, those factories would be designed. What you neglect to realize is that, when dealing with a broad hypothetical like 'non-designed' and 'designed' universes, it's entirely possible for such factories to exist and not be designed. They could simply exist, sans explanation. They could be the result, even the instantaneous result, of who-knows-what law of the universe.

I think you know this, which is why you keep comparing to 'NDE' rather than the 'undesigned universe'.

The statement of 'a designer could always have done things radically differently than we see' A) always applies, and B) applies to the hypothetical non-designed universe as well. In fact it's trivial to imagine a universe where species spontaneously generate from such and such, complete with hypotheses about how these exist brutely, or as the result of such and such law (indeed, both a design AND a !design universe can instantiate such), and the complaint that a designer would be able to create species in so many more ways - for instance, he could create a world of development over time and common descent.

So really, your comparison is fundamentally flawed. On a tangent, I'll point out another flaw with your statement here:

I think really smart designers can think up even more peculiar arrangements that are out of reach of NDE.

'These arrangements are out of the reach of NDE' is pretty much the core ID claim, from Dembski to Behe.

Doctor Logic said...

Crude,

If your comparison is between 'The general theory that the universe is designed' and 'The general theory that the universe is not designed', and you believe that the latter is 'NDE', while the former is some huge variety of options, you're wrong for the reasons I've pointed out already.

I've already explained that NDE competes against non-design alternatives as well.

NDE shows we were not designed because it beats both the design and non-design alternatives.

The only way specifying 'NDE' matters is if 'the general theory that the universe is not designed' cashes out to 'NDE, period'. If Karl actually knows 99 recipes, and among Bob's recipes is a recipe for burgers, then 'Someone is cooking burgers' isn't of much help in determining whether Karl or Bob is cooking.

Are there any non-design alternatives that predict a world like NDE would create?

I can't think of even one.

Okay, I've just read your latest comment wherein you suggest that the arrangement of species could be a brute fact, sans explanation.

I think this is awesome! And hilarious! Because the number of variations possible in the "brute facts" theory forms the largest set of all. It's the same size as the "any design will do" scenario. And NDE clearly beats them all from a Bayesian perspective.

The "all facts are brute" theory has the same infinitesimal likelihood for ANY scenario. The "any design will do" theory is the same. Yet NDE prefers descent, common descent, common composition, gradual change, generally increasing complexity over time, etc. NDE predicts an infinitesimal subset of the "all facts are brute" or "any design will do" theories.

What you have done is just explain that it is possible that NDE did not occur, and that it is merely a brute fact out of many possibilities that we got what we got.

This is EXACTLY like saying that a spin on the 1000-slot roulette wheel is just as good an explanation for a 4 as a spin on the 4-slot wheel.

Crude said...

One way to put the reply here, DL, is that you have to do something like the following:

Explain, given the incredible spread of possibilities that are subsumed under 'designed universe' and 'non-designed universe', why we should expect that a universe that is designed will not be a NDE universe (and even here that means, 'a universe where NDE is the most popular scientific theory'.)

You'll have to do so with the following caveats in mind.

* Recognizing that the spread of what's possible given 'non-designed universes' goes way, way, way beyond 'NDE'.

* Recognizing that your spread of designer *intentions* and *limitations* is incredibly broad, as they're undefined, while at the same time not only subsuming evolution as a design tool, but demonstrably doing so.

* Recognizing that NDE, as far as science is concerned, is silent with regards to whether evolution is guided in the relevant sense. There's no (I'm still waiting for that paper) scientific test of the 'God guides evolution' hypothesis.

* Recognizing that the Origin of Life is separate from the NDE question.

I'm not pointing out every possible objection - I'm even leaving aside the various objections (ID and non-ID) to NDE, or the fact that neo-Darwinism has proven to be extraordinarily plastic as a concept. I'm trying to keep the metaphysics as minimal as possible here. But I'm trying to illustrate just what hoops you have to jump through to get to your conclusion that NDE isn't something we should expect from a designer.

When you gave your Bob/Karl example, you suggested that Karl only has one recipe, and Bob has 99. But, I've pointed out, Karl has far more than one recipe - and both Karl and Bob have burger recipes.

ozero91 said...

I think William gets it.

Crude said...

I've already explained that NDE competes against non-design alternatives as well.

NDE shows we were not designed because it beats both the design and non-design alternatives.


NDE, as you posit it, is itself a non-design alternative.

Furthermore NDE, as far as science can tell us, is both a design and non-design option.

What you have done is just explain that it is possible that NDE did not occur, and that it is merely a brute fact out of many possibilities that we got what we got.

Did you completely miss the part where I was talking about hypothetical universes - ie, universes other than the one we currently exist in? In principle possibilities?

You seem to be under the impression here that I'm denying that NDE took place, or that denial of NDE is instrumental to my argument here. I'm not, nor is it. I've pointed out that NDE, as science posits it, is entirely compatible with design.

You seem to interpret this as some kind of skepticism on the line of, 'Okay, I know such and such evidence LOOKS like NDE is true - but in reality, NDE is false!' But no. I'm granting that the scientific theory of NDE is, as a matter of fact, true. I'm pointing out that it isn't the evidence which is compatible with design, but the fact of NDE itself, as far as science goes.

Now your responses to me start to make sense. It seems that you think people who believe in design are committed to *denying* that the scientific theory of NDE is true.

You're wrong.

ozero91 said...

Guys, I think he's arguing, from an epistemic position, that unguided evolution is far more probable than guided evolution. Though I still think William has a point.

Crude said...

ozero,

Guys, I think he's arguing, from an epistemic position, that unguided evolution is far more probable than guided evolution. Though I still think William has a point.

I don't think so. I think he really is under the impression that a theist must *deny* the scientific theory of evolution, and that NDE as a scientific theory is incompatible with a designer. The *evidence* for NDE is what he thinks is compatible with theism, in a conspiracy theory sense.

If I'm wrong, he'll correct me, but I think I finally understand just why there's so much conversation going, and so little progress is being made.

Crude said...

I think he really is under the impression that a theist must *deny* the scientific theory of evolution,

Amending: a theist who believes God designed, etc.

ozero91 said...

I think this is what he's arguing:

Assume that God has an infinite or at least a massive number of methods by which he can produce a diveristy of life on our universe. (This includes evolution). Assume that we do not know God's will. Assume that the only way a diversity of life could appear in our universe, without the aid of God, is abiogenesis and evolution. (Or even if abiogenesis and evolution are not the only methods, the number of options is still vanishingly small compared to the number of ways God could create a diversity of life in our universe). Apply Bayes theorem, and the results show that our presence is more likely explained by unguided evolution than guided evolution, regardless of whether or not God exists. One problem though is that this does not take the will of God into account, because God is obviously different from a roulette wheel with an infinite number of slots.

Crude said...

ozero,

Assume that the only way a diversity of life could appear in our universe, without the aid of God, is abiogenesis and evolution. (Or even if abiogenesis and evolution are not the only methods, the number of options is still vanishingly small compared to the number of ways God could create a diversity of life in our universe).

Roughly, that's what I thought his argument was at first as well - and that's what I've been arguing from comment 1. The idea that 'if there is no designer of life/universes/etc, then abiogenesis/NDE is the only game in town for the diversity of life, etc'. Except that fails: when you're considering the ways life, etc, count exist in hypothetical universes, with the only requirement being 'not designed', you're off into the land of tremendous, ridiculous numbers of possibilities. Various theories are, upon investigation and taking on such and such ground assumptions for the sake of science, etc, more likely in *our specific universe*, but not in universes considered in general.

The point there is that a 'non-designed universe' is compatible with many, many, many scenarios where life and diversity obtains, with NDE not being required or not playing a major role. Likewise, a 'designed universe' is compatible with many, many, many scenarios where the same obtains, including universes where NDE is the primary vehicle. Remember, the sole requirement is that these things not be designed, and they be logically possible. That leaves you with one heck of a lot of ways a universe with life could exist, in principle.

I think there's another aspect to DL's reasoning, and now I'm asking him about it. We'll see what he says - no need to speculate too much, the guy's right here. But if I'm right, it could explain why there's been little progress here.

William said...

ozero:

"Assume that God has an infinite or at least a massive number of methods by which he can produce a diveristy of life on our universe. (This includes evolution). Assume that we do not know God's will. Assume that the only way a diversity of life could appear in our universe, without the aid of God, is abiogenesis and evolution. (Or even if abiogenesis and evolution are not the only methods, the number of options is still vanishingly small compared to the number of ways God could create a diversity of life in our universe). Apply Bayes theorem, and the results show that our presence is more likely explained by unguided evolution than guided evolution, regardless of whether or not God exists. One problem though is that this does not take the will of God into account"

Yes, I completely agree, you said better than I did. Though I would put "creation by God" in place of "will of God" above, which brings the circular dependency problem (also shown by the causal map) into better relief.

Sammy Onomato said...

B. Prokop wrote:
“Sammy, yes I would object to such an assertion - and strongly. "Science" has no business weighing in on the matter at all. That is precisely what is meant by your statement "Scientists should be free set up their scientific models or explanations without regard to what any particular religious beliefs people may have." Once a person starts claiming that an observed event is either guided or unguided, he is no longer remaining neutral vis a vis "religious beliefs people may have".

Surely that can be agreed upon?”


I wasn’t saying science should remain neutral. Rather it should not worry about any theological or metaphysical beliefs that any particular religion adheres to when seeking to explain natural phenomenon.

Atomic decay is according to science an unguided process. The melting of ice according to science is an unguided process. The motion of the planets around the sun is according to science an unguided process. You may find that offensive because it goes counter to your religious beliefs, but science should not have to take your religious beliefs (or mine, for that matter) into consideration when explaining how those things work in the world.


Crude said...

Atomic decay is according to science an unguided process. The melting of ice according to science is an unguided process. The motion of the planets around the sun is according to science an unguided process.

Not in the relevant sense of the word 'unguided', no - not for any of those things. Science has zero, nada, nein ability to determined whether such things are guided or unguided with regards to God, etc.

Now, that may cause some people severe distress. They may want, and want desperately, that science be able to discover certain things are true or false. But all their hopes in the world won't change the limitation.

See, it's not just the metaphysical beliefs of religion that science couldn't give a whit about, but an atheist's or naturalist's metaphysics as well.

Doctor Logic said...

ozero91,

Assume that God has an infinite or at least a massive number of methods by which he can produce a diveristy of life on our universe. (This includes evolution). Assume that we do not know God's will. Assume that the only way a diversity of life could appear in our universe, without the aid of God, is abiogenesis and evolution. (Or even if abiogenesis and evolution are not the only methods, the number of options is still vanishingly small compared to the number of ways God could create a diversity of life in our universe). Apply Bayes theorem, and the results show that our presence is more likely explained by unguided evolution than guided evolution, regardless of whether or not God exists.

You have this correct.

One problem though is that this does not take the will of God into account, because God is obviously different from a roulette wheel with an infinite number of slots.

You're saying that a specific, predictive theory of a designer will predict a non-uniform distribution of fauna. I totally agree. But that's not what ID does. Their goal has been to claim that we need say nothing about the designer.

This is analogous to claiming we can infer naturalism without ever talking about a specific set of predictive theories. Bayes says it ain't so.

ozero91 said...

"Atomic decay is according to science an unguided process. The melting of ice according to science is an unguided process. The motion of the planets around the sun is according to science an unguided process. You may find that offensive because it goes counter to your religious beliefs, but science should not have to take your religious beliefs (or mine, for that matter) into consideration when explaining how those things work in the world."

You'll have to ask Bob to clarify what he means by "guided." Maybe he is speaking in terms of final causality, like the tendency of an unstable nucleus to emit an alpha particle instead of just a proton, or the tendency of a water molecule to move into the liquid or gas phase when heat energy is added to it, instead of turning into photons. He may be arguing that these regularities are only possible if God is there to sustain them (like the Fifth Way, I suppose). This is something that science has no say on, the hypothesis "If God does not exist, then there would be no final causes" cannot be settled in a lab.

Crude said...

William & Ozero,

I'd appreciate it if one of you could give some input on what I'm saying here, if you guys have the time and interest.

The claim here seems to be that, while in a designed universe life could come about in any number of ways, in a non-designed universe life could only come about in a very particular way or set of ways - NDE, here. Therefore, if we find ourselves in an NDE universe, we should (putting all other considerations aside) expect we're in a non-designed universe.

My reply is that this is incorrect. 'Non-designed universe' and 'designed universe' are extraordinarily broad, arguably unlimited - and the in principle ways life could exist in a non-designed universe are phenomenal, and not so limited.

At the same time, an NDE universe/universes where evolution takes place, are also within the space of a designed universe's possibilities.

So it follows from these broad facts, taken as they are, that finding ourselves in a universe where NDE (or some evolutionary process) is the means by which life had propagated, should not lead one to infer they live in a non-designed or designed universe.

Do either of you see a problem with the reasoning as I laid it out?

On a second point, I'd point out that the ID proponents are going to deny that they live in an NDE universe anyway. They believe that design can, in fact, be inferred (scientifically, no less) by the data we have onhand, though the extend to which they question NDE differs among them. (Some deny common descent, others accept common descent but deny NDE can reasonably accomplish X or Y, etc.)

So, from the ID view, whether we live in a universe where NDE is true is immediately questioned already.

Any problems with what I've said?

ozero91 said...

"My reply is that this is incorrect. 'Non-designed universe' and 'designed universe' are extraordinarily broad, arguably unlimited - and the in principle ways life could exist in a non-designed universe are phenomenal, and not so limited."

Hi Crude, I think I'll have to continue this discussion tomorrow, since I have to stop procrastinating and finish a paper. But I think he is arguing solely based on the cosmological parameters of OUR actual universe, which greatly limits the number of plausible naturalistic theories. DL may be assuming that there is no multiverse for the sake of his calculation, but you'll have to ask him. Once we bring in the multiverse, I think probabilities become insane, as with your exchange with Angra.

Crude said...

Another way to explain what I'm saying is this.

Consider the questions "What sort of things could I see in a universe that was designed?" and "What sort of things could I see in a universe that was not designed?"

It seems that the overlap between those questions, taken as they are, is considerable. And notice, I'm not talking about, say... 'this universe that I'm already in' with that question. I'm talking about hypothetical universes - ones that could have completely different types of laws, sets of laws, etc. It seems that both of those questions would have a tremendous number of possibilities available, while at the same time there would be a tremendous amount of overlap. It's, to use the Bob and Karl case, going to yield an answer (with those terms considered alone) of Bob and Karl each having a billion recipes, and 99%, possibly 100%, of those recipes are in common between the two. Rather than a situation where Bob has only 1 possibility and Karl has a billion, and it includes Bob's.

So, since the spread is that large, and the overlap is that large, an NDE universe does not shift you to 'an undesigned universe' based on the bare considerations you're working with. Nor is it going to shift you to 'a designed universe'.

Now, you can bring in additional metaphysical stipulations - maybe argue that (a designing) God is metaphysically necessary given such and such axioms, etc. Or you can do other things. But at that point, what will be doing the work is not merely 'designed universes v undesigned universes', but the additional arguments and concepts.

Doctor Logic said...

Crude,

I don't think so. I think he really is under the impression that a theist must *deny* the scientific theory of evolution, and that NDE as a scientific theory is incompatible with a designer. The *evidence* for NDE is what he thinks is compatible with theism, in a conspiracy theory sense.

The ID crowd has essentially been denying that NDE is the explanation for speciation because they claim NDE is inadequate to the task. They say NDE could not account for the life we see in an unguided scenario because NDE cannot create complex life all by itself.

I think you're referring above to theistic evolutionists (TEs) who accept NDE as an explanation for speciation in an unguided scenario, but they assert that NDE was the tool chosen by God to implement speciation. Basically, TEs are saying that NDE is adequate to explain speciation (even in unguided worlds), but redundant in our world in the sense that it was optional from God's point of view.

My argument is that NDE is not only redundant from a theistic point of view, but that it represents a tiny sliver of the possible ways a designer could implement life. Meanwhile, there is no known competitor to NDE in the wholly unguided scenario.

Crude, you've denied this last statement, and argued that just as many life-bearing worlds are possible given non-design as in the case where there is design. You've supported your claim with the idea that the existence of species could be brute facts (the equivalent of species-specific physical laws).

If you were serious about this, you would have to conclude that Paley's watch was no less probable under non-design than under design. For it could be that there is a physical law that such a watch exist at that time and place it was found. The same would apply within any ID argument. Bacterial flagellum? There's a law for that!

No, it must be that we're considering paths that life took, not merely destinations. The only path solution we know of for unguided scenarios is NDE. But when it comes to design, the paths need not be visible at all. Path is unimportant to design. There need be no apparent natural history at all.

William said...

Crude:

' Consider the questions "What sort of things could I see in a universe that was designed?" and "What sort of things could I see in a universe that was not designed?" '

It's my conceit that there might be as many kinds of life as there are solar systems containing life, which I hope is many more than just us.

This would mean that the probability of evolution causing life, via own local DNA-based modern synthesis definition of evolution, is low overall. And in that case, maybe every one of Doctor Logic's barbecue recipes gets cooked out there somewhere.

The probabilities of design would vary based on those assumptions too. It's inscrutable as it stands.

Crude said...

ozero,

But I think he is arguing solely based on the cosmological parameters of OUR actual universe, which greatly limits the number of plausible naturalistic theories.

No rush at all, thanks for the reply.

I think there are big problems with that kind of response.

Sure, once you come up with observations and descriptions of a particular universe, you've limited the space of your possibilities. The problem is, you've limited both your designed AND undesigned possibilities. If you're in a universe where spontaneous generation is true and NDE is false, that's that. But it makes little sense to go on and say that given you're in such a universe, clearly it's not designed on the grounds that God could have conceivably used NDE or who knows whatever number of variants of processes to achieve the results, but He didn't. You could also say that clearly the universe wasn't made without design, because hey, the space of non-designed possibilities is vastly greater than what we see. On their own terms, they both make about as much sense.

It's not really about 'bringing in the multiverse' here. It's about what we should hypothetically expect given the bare qualifications of 'design' and '!design'. NDE, meanwhile, is compatible with both design and !design - and the hypothetical !design possibilities go radically NDE anyway.

Crude said...

The ID crowd has essentially been denying that NDE is the explanation for speciation because they claim NDE is inadequate to the task. They say NDE could not account for the life we see in an unguided scenario because NDE cannot create complex life all by itself.

It's a little more nuanced than that, but for our purposes yes - I've said as much during this exchange. ID proponents question the truth of NDE, and point at what they consider to be tremendous flaws in the NDE theory.

I think you're referring above to theistic evolutionists (TEs) who accept NDE as an explanation for speciation in an unguided scenario, but they assert that NDE was the tool chosen by God to implement speciation. Basically, TEs are saying that NDE is adequate to explain speciation (even in unguided worlds), but redundant in our world in the sense that it was optional from God's point of view.

Plenty of TEs would question and even reject that evolution is 'unguided' in the relevant sense of the term, especially where science is considered. More than a few atheists would object to the claim that NDE, as far as science is concerned, is unguided. That gets into the land of the philosophical and metaphysical.

My argument is that NDE is not only redundant from a theistic point of view, but that it represents a tiny sliver of the possible ways a designer could implement life. Meanwhile, there is no known competitor to NDE in the wholly unguided scenario.

There are plenty of competitors in principle. It's redundant from a non-designed point of view as well(and by the way, there you go, talking theism again.)

So no, it's not a tiny sliver - remember, we're talking about the terms of in-principle possibilities. When you talk about the list of designer possibilities, you're in the realm of hypotheticals. But that's exactly the same realm you have to go to with the non-designed possibilities to make the argument you're making here.

And the moment you do, your argument folds.

Crude said...

Crude, you've denied this last statement, and argued that just as many life-bearing worlds are possible given non-design as in the case where there is design. You've supported your claim with the idea that the existence of species could be brute facts (the equivalent of species-specific physical laws).

I've supported my claim with a variety of other possible scenarios too: differing fundamental laws of the universe, past-eternal species, spontaneous generation and corresponding laws, and more.

If you were serious about this, you would have to conclude that Paley's watch was no less probable under non-design than under design. For it could be that there is a physical law that such a watch exist at that time and place it was found. The same would apply within any ID argument. Bacterial flagellum? There's a law for that!

The funny thing is, that's actually fine from the ID perspective. The ID view is not 'the only way these things could come to pass is by a miracle' or even 'the only way these things could come to pass is by design!' It's a conditional inference based on the data onhand: 'the evidence, as it stands, is that NDE is incapable of achieving X'. ID proponents openly say that it's possible new data /discoveries could overturn their inferences, so 'there's a law/explanation/whatever!' for that' isn't a dealbreaker from the ID perspective - if it comes to pass, that's that.

What's more, I've pointed out that what I'm saying obtains if you only consider - and I've explicitly minimized the metaphysical and axiomatic assumptions aside - the bare question of 'designed universe' versus 'undesigned universe'. Start taking on metaphysics and axioms, and things change. Indeed, you may be able to circumvent Paley entirely, which I think is actually the case.

But when it comes to design, the paths need not be visible at all.

Paths don't need to be visible for !design either, especially if visible means 'discoverable by humans'. In fact, arguably there's less of a restraint on !designed situations - there must always be some kind of path, even a circuitous one, with a design scenario. 'At least one mind comprehends the situation.' With !design? There's no need for any mind to apprehend things. It can, indeed, but unapprehendable.

Crude said...

William,

Well, at the very least I can say that's a very interesting response.

grodrigues said...

Argument 1: wherein the Author shows that if you choose your probability measure just right you can get what you want.

Let X be the set of all possible universe histories, X_d the subset of universe histories where life was designed and X_u its complement, that is, the set of universe histories where life was not designed. Now take the principal filter F generated by X_d. Trivially, X_u is not a member of F. By the Boolean prime ideal theorem there is an ultrafilter G extending F and not containing X_u. Given the ultrafilter G, define a function p on the power set Boolean algebra of X in the following way:

p(E) = 1 if E is in G, 0 otherwise (E contained in X)

An easy computation shows that p is a probability measure (sigma-additive and spectral). By construction p(X_d) = 1 and p(X_u) = 0.

I think my probability measure is a very nice one; it gives me just the result I want. Granted, it is a tad artificial, but at least it *is* a bona-fide probability measure. Show me yours.

Argument 2: wherein the Author borrows the Ultimate Design Argument from someone who he will not name because he has already forgotten his identity.

Consider now the set of Y of all possible universe histories that unfold *without* direct intervention of The Deity. For the sake of argument grant everything that atheists want to say about the appearance and evolution of life. The Deity being omniscient knows all of the histories y in Y and knows in particular which ones will have life in a little blue planet called Earth, and then the ones in which life evolves to the stage of there actually being a rational species, and then the ones where said species develops and unfolds until it invents a marvelous thing called internet and blogs, in which a combox poster takes on the mantle of logic snicker and posts an "argument" against design. Denote such a history by y_0. The Deity, being the Deity and thus omnipotent and outside of Time, can choose among all the possible histories y in Y, precisely the history y_0 to actualize. So in this scenario, we have a deliberate choice of universe history in which there is life but no direct intervention for said life to arise. So in this scenario there is design (e.g. choice of a particular universe history with a particular outcome) without design (no direct divine intervention, just the regular unfolding of the laws of physics just like your average skeptic atheist likes).

note: yes, I am bored. But then the whole "argument" is dreadfully boring, vacuous and irrelevant.

B. Prokop said...

"You may find that offensive because it goes counter to your religious beliefs"

Believe me, it takes far more than that to offend me!

Ozero91 understood what I meant. In The Book of Common Prayer we find the words "Almighty God, we bless thee for our creation [and] preservation". The word preservation refers to God's continuing sustainment of creation at each moment of time. That would include guiding the movements of the planets, and ensuring that an ice cube will melt at the proper temperature.

Sammy Onomato said...
This comment has been removed by the author.
Sammy Onomato said...

B. Prokop wrote:
The word preservation refers to God's continuing sustainment of creation at each moment of time. That would include guiding the movements of the planets, and ensuring that an ice cube will melt at the proper temperature.

Thanks for explaining that.
But why, then, would you object to science explaining natural phenomenon such as atomic decay or biological evolution as being unguided?

B. Prokop said...

"But why, then, would you object to science explaining natural phenomenon such as atomic decay or biological evolution as being unguided?"

Because that's a philosophical/theological statement, and not a scientific one. I have zero objection to you or anyone else asserting that the processes are unguided, as long as you understand that.

Doctor Logic said...

William,

This would mean that the probability of evolution causing life, via own local DNA-based modern synthesis definition of evolution, is low overall. And in that case, maybe every one of Doctor Logic's barbecue recipes gets cooked out there somewhere.

Groovy.

The probabilities of design would vary based on those assumptions too. It's inscrutable as it stands.

But, for every naturally evolving world, aren't there vast numbers of ways that God could tinker in obviously visible ways that natural evolution cannot replicate? Precambrian rabbits? Ghosts? Plastic deer?

I'm not asking you to put a precise number on the number of ways. Just some lower bound. I did this earlier, and came up with at least 2^5N where N is the number of species that have ever lived. I don't have to think this is at all precise. We can imagine worlds where N is very small, though under NDE we expect N to be quite large when there are intelligent species. It doesn't really matter, even if you think our world is not representative, and that N is typically low (like 100), the case against design is still devastating.

B. Prokop said...

Let's put this another way. If you say "Evolution is unguided" is a scientific statement, then so also are "The unexamined life is not worth living," "The fault, dear Brutus is not in our stars, but in ourselves," or "Love your neighbor as yourself."

B. Prokop said...

Doctor "Logic",

By your reasoning, the case against your own existence is "devastating", since there are trillions upon trillions of DNA combinations that could have resulted in you being somebody else, but only one in you being you. Therefore, I must conclude that the probability of your existence is vanishingly small, and should disbelieve in your existence.

Doctor Logic said...

Crude,

[ID is] a conditional inference based on the data onhand: 'the evidence, as it stands, is that NDE is incapable of achieving X'. ID proponents openly say that it's possible new data /discoveries could overturn their inferences, so 'there's a law/explanation/whatever!' for that' isn't a dealbreaker from the ID perspective - if it comes to pass, that's that.

I find this claim completely disingenuous (on the part of ID proponents).

ID proponents spend their time pushing arguments that are the equivalent of "Aerodynamics says the bumblebee cannot fly!"

As if the whole of physics, chemistry and biology will collapse in the face of a mystery that hasn't yet been studied in detail. If I make a naive calculation from a position of ignorance, I will often find that some things seem impossible. This is expected under NDE.

Suppose I said "There's no way a gecko's feet could cling to glass without them being so sticky as to preclude movement!"

You would say, "Well, duh, if you assume its feet stick using some sort of glue, then you're right, but why assume this? Why not actually see how it works?"

In the case of NDE, the actual history of how a feature evolved is often (and predictably) obscured by time - the loss of intermediate species, the distorting effect of mutations, etc. ID proponents ignore this. They take naive assumptions, and ignore the accumulated evidence for NDE as if none of the prior data has any bearing on the likelihood of gaps needing to be filled by a designer.

It doesn't take a genius to know that, given the success of our scientific models to date, the bumblebee can probably fly aerodynamically, without any special laws or divine interventions.

A priori, if the world were patched up with special cases or divine interventions, then we would not have expected our laws to do as well as they have done. There are far more patches and interventions that could be made that don't fit an NDE picture than that do. And that's the inference I'm talking about.

B. Prokop said...

"if the world were patched up with special cases or divine interventions"

That is not how a Christian would view God's activity in our world - not at all. It's not even a poor characterization of it - it's pure gobbledygook.

First of all, from the Christian perspective there is no "patching up" - there is Creation and Preservation, all very orderly.

Secondly (as I have posted many times on this website), any miraculous event must relate in some fashion to the Incarnation and Redemption. All of them. They are not haphazardly strewn across history, but all quite purposeful, highly directed, and clearly recognizable as being apart from the natural order.

When Christ walked on the water, the disciples did not ask "How is He doing this?" or "What are the probabilities against this happening?" They asked "Who is this person?" And that dear Dr. Logic, is where you have fundamentally erred. You are asking the wrong question. As long as you continue along such lines, you will never arrive at The Answer.

ozero91 said...

"That is not how a Christian would view God's activity in our world - not at all. It's not even a poor characterization of it - it's pure gobbledygook."

I agree Bob, this conception is pretty disastrous, but it seems to permeate neo-theism and the ID movement. Some people just think of God in an odd way, like a guy who sits in the cosmos and says "Okay, I'll mutate this now, and mutate this then, oh wait if I don't mutate this gene I won't get humans, better do that" etc.

B. Prokop said...
This comment has been removed by the author.
B. Prokop said...

Ozero91,

Thanks for that illustration. I've never cared for the ID movement, despite the fact I myself am a believer in design - I just don't regard the concept as "scientific" (as explained above). And concerning this particular thread, whenever an atheist starts making up his own definitions for theistic terms and concepts, it resembles a person walking in on a debate on the relative merits of paper or plastic, and starts going on about butterfly nets. Totally beside the point, and doesn't advance the conversation one whit.

"Dr. Logic", if you're going to argue with theists, at least make an effort to understand just what it is you are arguing against! As it is, you are firing artillery into the void, and we're all over here watching you engage in combat against an imaginary foe.

BenYachov said...

DL,

Just to do a little hit and run.

Bob is right & you are just going to have to except it.

Your must argue with the Deity your opponents believe in not the one you wished they believed in. Just because you are only competent in arguing against a narrowly specific god-concept doesn't grant you leave to equivocate and apply those argument to different god-concepts.

BenYachov said...

In the end it all comes down too the Classic Theism of Catholics, Eastern Orthodox, older Protestantism and Orthodox Jews and their philosophical assumptions vs the neo-Paley Theistic Personalist "deity" of post enlightenment modern philosophy.

ozero91 said...

"You're saying that a specific, predictive theory of a designer will predict a non-uniform distribution of fauna. I totally agree. But that's not what ID does. Their goal has been to claim that we need say nothing about the designer."

From this, I think DL made it clear that he was going after ID, and we all know how the ID'ers think. He's objecting to the vagueness in their "scientific hypothesis."

ozero91 said...

In addition, why start the probability from our universe, why not take the big bang into account?

Crude said...

ID proponents spend their time pushing arguments that are the equivalent of "Aerodynamics says the bumblebee cannot fly!"

Yeah, I don't find this to be an apt comparison whatsoever. It's far more on the mark to suggest ID proponents are saying 'You know, maybe string theory is wrong.'

By the way, just to get this out of the way: I don't think ID is science. I also don't think ID's opposite is science. I think ID asks some valid questions, and does make some valid criticisms. But I think when it comes to inferences about design or its lack, it's just another field than the one they want it to be.

In the case of NDE, the actual history of how a feature evolved is often (and predictably) obscured by time - the loss of intermediate species, the distorting effect of mutations, etc. ID proponents ignore this.

Ignore it? They can't shut up about it. It's an arrow in their quiver. Just as they can't shut up about Lenski's experiments and what's accomplished/observed by them.

Frankly, I think the response of, 'Sure, we don't have any real evidence of the proposed mechanisms being capable of accomplishing what we think they are. We can't be expected to have observations or experiments of such. But that's okay, because the theory says those things are real obscure!' is all that compelling. Even if you were to put design considerations entirely to one side, it would seem that that would support agnosticism on the adequacy of NDE mechanisms.

It doesn't take a genius to know that, given the success of our scientific models to date, the bumblebee can probably fly aerodynamically, without any special laws or divine interventions.

Look, classical physics has worked fine for over a hundred years. Just because some phenomena currently seem confusing according to the model is no reason to think some new kind of physics is needed.

or

Look, there's no problem with our theories of astronomy. See? Add a few epicycles and voila, we're good. Why's everyone getting so worked up about this? Show me a problem that can't be solved by adding epicycles!

That goes some way towards illustrating my skepticism here, even as a lifelong TE. Keep in mind, by the way, that the skepticism is in no way parasitic on a design inference. You can put the question of design entirely to one side and still have cause to be skeptical of the adequacy of NDE.

B. Prokop said...

Crude,

I'm glad that you brought up the Ptolemaic System, because most people don't realize how extensively we use that system even today. After all, we all (even astronomers) say "The sun will rise at 7:17 tomorrow." Nobody says, "By 7:17 tomorrow morning, the Earth will have rotated sufficiently to allow us to observe the Sun from our line of longitude", despite that being the more accurate statement.

(Apollo 11 astronaut Michael Collins once wrote that it wasn't until after his Moon flight that he finally started thinking about such things correctly. "No longer do I drive down a highway and wish the blinding sun would set; instead I wish we could speed up our rotation a bit and swing round into the shadows more quickly." Just goes to show that in most cases, and especially in those where it really counts, experience trumps book knowledge.)

William said...

DL:

Since we are talking competing belief systems here, competing speculations will not suffice. In the words of N. D. Mermin, shut up and calculate.

Let's streamline the causal model I gave you before by assuming that:
1. Evolution is almost all that influences causality in the model so that the direct influence of creation outside of evolution and spontaneous generation is nil compared with the influence of E on LHN. (Factually, this is not quite true, since we have origins of life prior to evolution).

2. Let's in turn make evolution the only way that LHN gets that way, so P(LHN | E ) = P(LHN).

This leaves C -> PPU -> LHN and C -> LHN as the only causal pathways of any significance.

So, let's look at the extremes of the Bayesian priors.

If P(C) = 0, then this reduces to PPU -> LHN, which is what you want (NDE, no need for a God for life at all). But if P(C) = 1, then it's just a question of whether God set up evolution at the start or whether post-abiogenesis, prehistoric interventions also occurred ( C -> PPU -> LHN plus or minus C -> LHN ).

William said...

DL:

Continuing the prior note, since

C -> PPC -> E -> LHN is the causal chain without the tinkering post-abiogenesis and C -> E -> LHN is the causal chain with special tinkering, we cannot use your argument to decrease the probablility of C, since C is at the base of both causal chains!

We _can_ use external evidence for the C -> E link (without the PPU mediator), if any, to increase the probability of C. But that evidence winds up simply being more potential external evidence for C, which is the circularity problem again :-).

Doctor Logic said...

Bob,

By your reasoning, the case against your own existence is "devastating", since there are trillions upon trillions of DNA combinations that could have resulted in you being somebody else, but only one in you being you. Therefore, I must conclude that the probability of your existence is vanishingly small, and should disbelieve in your existence.

While it may be a priori highly improbable that I have the DNA sequence that I actually have, there are equally slim odds that a genomic sequencing would find my DNA sequence to be what it actually is. The two cancel out.

If I deal you a 3, 7, 2, K, J of clubs in poker, that specific hand in that specific order is extremely improbable. But you can convince yourself that it is true by looking at your hand. The odds cancel out because the odds of you reading that hand when that isn't your hand is equally low.

So, the debate here isn't about affairs being patently how we observe them to be. What we're talking about are models that predict the class of observations we're making.

Under NDE, there's a predicted connection between my DNA and the DNA of my parents and of my children. Agreed?

If no NDE, then there need be no connection. It's easy to imagine a counterfactual world, say, where conception involves God implanting an embryo into every living being, and that each embryo has its very own genome with very little in common with anybody else's. Agreed?

Or even if God doesn't do this every time there is a birth, he could do it when he wants to create a new genus or species. Foxes would have no genome in common with mice, say.

So, it is very easy to imagine designed counterfactuals to our world. You gave an example - why do I have my DNA and not yours, and vice versa? If our respective genomes were swapped, that's a counterfactual, but a counterfactual that is compatible with NDE.

However if my genome had nothing in common with my family's genome, that would contradict NDE.

So, let's estimate the number of counterfactuals of each variety. The number of NDE-incompatible counterfactual worlds is vastly greater than the number of NDE-compatible counterfatual worlds. For every NDE-compatible counterfactual world, the number of designed counterfactuals is huge. This is why NDE is confirmed by the evidence.

What theists are saying is that even though NDE is capable of creating the life we see, it was designed, despite the fact that design is compatible with non-NDE, and physics isn't.

Crude said...

What theists are saying is that even though NDE is capable of creating the life we see, it was designed, despite the fact that design is compatible with non-NDE, and physics isn't.

...And !design is compatible with non-NDE as well. Saying that 'physics' isn't compatible with non-NDE is a red herring, since whatever actually took place is almost tautologically 'compatible with physics'. That large, large spread of designed and undesigned worlds is not a huge collection of 'worlds where things take place that are incompatible with physics'. They're worlds with different physics - and there's nothing about !design that requires there be even consistent physics within a single world.

And, again, putting aside considerations of metaphysics, etc, 'non-design' is compatible with just about any arrangement you can think up. The problem's compounded when you realize that classical theism isn't required for design either - 'natural designing agents' themselves occupy a huge spread of the possibilities.

You keep wanting to play hypothetical possibilities that are in play prior to observing a world off against a particular observed world in a way that simply doesn't work.

Crude said...

Also,

So, let's estimate the number of counterfactuals of each variety. The number of NDE-incompatible counterfactual worlds is vastly greater than the number of NDE-compatible counterfatual worlds. For every NDE-compatible counterfactual world, the number of designed counterfactuals is huge. This is why NDE is confirmed by the evidence.

You also have to keep in mind the level of confirmation versus the variety of counterfactuals out there. A universe where a designer, say... intervened in the bacterial flagella, and nothing else, would I suppose count as an 'NDE-incompatible world'. It also happens to be a world we'd have severe trouble detecting in a tremendous number of cases in. Considering that you don't even need to directly alter biology to dramatically intervene in evolutionary results, that's saying something.

It gets worse once you realize the actual levels of confirmation for NDE (as opposed to evolution generall), the plasticity of NDE as an idea, and more.

Again, you're going about this the wrong way just because you fail to recognize the sheer spread of in-principle possible !design options sans metaphysics, assumptions, etc. But I think you're not even appreciating just what's involved with the spread of counterfactual NDE worlds.

Doctor Logic said...

William,

Here's my objection. If you only consider LHN (i.e., exactly the observed species and genomes), then there's nothing to say. However, that's not what we do when we compare theories. When we compare theories, we ask whether how our observations compare with the predictions of our theory versus its competitors, i.e., with counterfactuals.

So, in the following, I'll substitute L for LHN. We're not only interested in the life we observe but in any sort of life.

My inference has nothing to say about C->PPU->E->L in the case where the creator doesn't value L.

A god could create a billion universes for any number of arbitrary reasons. Suppose a god only cares about the distributions of stars in these universes, and doesn't care about any life that is created as a byproduct. Of course, this isn't the god of Christianity, but it's a conceivable god/designer. Such a god did not design life on our world any more than we designed exhaust emissions. We are not a factor in this god's plan. The deist god doesn't give a 4@%& about life.

With such a god, the only known way to get life is NDE (Crude's objections notwithstanding).

On the other hand, the C->E->L implies that L is part of god's design. Now, what does god value in this case? If it's life, then, all things being equal, C->E->L competes against C->L.

C->L represents trillions more ways to create life (to be conservative) than C->E->L does or C->PPU->E->L.

I'll say it again. E is extremely restrictive in what it can do. All things being equal, what are the odds that a god bothers to utilize E in creating life? Roughly, we would divide the volume of ways described by C->PPU->E->L and C->E->L by the number of ways of achieving C->L. That number is so tiny that it deserves the name infinitesimal.

That's why NDE is overwhelming evidence that gods, if they exist at all, didn't design life.

William said...

DL:

I guess I can't help you.

The problem is that an agent, even in our human experience of animals alone, automatically narrows their possibilities of action by preferences that are intrinsic to the agent, in a way that is not merely based on what we think is possible for that agent.

So, the causality of what you call "the odds that a god bothers" to do things in a particular way cannot be used by me as part of my model, since I can't imagine how to model the causality of a godlike psychology. Sorry, that level of causal modeling's out of my league.

B. Prokop said...

Dr. Logic's conception of the probabilities is ass-backwards. Let's use my breakfast this morning as an example. Prior to 6:30 today, it was possible for me to have a breakfast of ham and eggs, or cereal, or toast and jam, or pancakes, or... you get the idea. the probability of my choosing one over the other is a meaningful figure. However, after 7 AM, when I'm sitting down to my bowl of cereal (as it turned out), any such figures are completely irrelevant. The probability of my having eaten cereal for breakfast today is now 1.

The same goes for the universe we live in. There may (but I still don't think so) be some small value in speculating how many different possible universes could have been created. But in 2012 A.D. that no longer matters. The probability of a universe like ours being created is now 1.

This thread has gone from beyond silly (which is where it began) to completely incoherent. If it proceeds further, I suspect we will descend to un-talk (a word I just coined, with apologies to C.S. Lewis - see Perelandra)

SteveK said...

Is DL an IDist now that design theory can be tested, or is DL talking about philosophy and using math as a cover?

Crude said...

On the other hand, the C->E->L implies that L is part of god's design. Now, what does god value in this case? If it's life, then, all things being equal, C->E->L competes against C->L.

Even on this point, it's more complicated than you're appreciating - C->E->L and C->L are not mutually exclusive options. There can be mixtures.

Worse, along the lines of what William just said - once you start talking about what a God values, you're off into some interesting territory. Even biblically, God valued a lot more than life. A God can value multiple things. A God can value one thing for X along of time, and then change his mind. (Before Ben comes at me with an internet switchblade, I'm talking about the sort of god ID is associated with.)

Hell, a god may even have a use for life that's purely practical. If God values paintings, not life, you still may have life showing up to create paintings.

I've put forward other criticisms here, but once you start getting God-psychology into the mix, you're in an even worse situation than you started as a skeptic.

Doctor Logic said...

Crude,

I've put forward other criticisms here, but once you start getting God-psychology into the mix, you're in an even worse situation than you started as a skeptic.

Why?

I don't see a consideration of possible psychologies decreasing the number of conceivable life-bearing worlds that a god might create.

Crude said...

I don't see a consideration of possible psychologies decreasing the number of conceivable life-bearing worlds that a god might create.

Sure they will, because psychology encompasses more than a single desire for 'life'.

Doctor Logic said...

Bob,

However, after 7 AM, when I'm sitting down to my bowl of cereal (as it turned out), any such figures are completely irrelevant. The probability of my having eaten cereal for breakfast today is now 1.

And, if I were challenging the factual observations of biology, you would have a point. But that's not what I am doing. I am making an inference from observations to a theory.

Let's go back to the BBQ pit. If you can cook 99 recipes, and Karl only hamburgers, and we discover that hamburgers were cooked, then it's much more likely that Karl was the cook than that you were.

In this inference, I'm not challenging the observed data. Yes, in this puzzle, it is assumed that hamburgers were cooked with 100% confidence. The question is, who was most likely the cook?

An omniscient being would know that either you were the cook or Karl was the cook with 100% probability. But we're not omniscient, so we play the odds. Playing this game 198 times would reveal to us that, all things being equal, there is just a 1% chance that you cooked the hamburgers in question.

Doctor Logic said...

Crude,

I don't see a consideration of possible psychologies decreasing the number of conceivable life-bearing worlds that a god might create.

Sure they will, because psychology encompasses more than a single desire for 'life'.


Some psychologies will. Some will not. So what? How does that help you? How does it shift the distribution of worlds that a god would create from plainly non-NDE worlds to plainly-NDE worlds?

NDE just makes life, not any particular kind of life. If there are more variations of life possible with non-NDE design, why rely exclusively on NDE?

A lot of fiction involves designed or non-NDE worlds. Non-NDE worlds are more interesting to us because more is possible in them. I don't see how you're going to break my inference with psychology. You would have to show that most psychologies that want life want NDE life.

Doctor Logic said...

William,

The problem is that an agent, even in our human experience of animals alone, automatically narrows their possibilities of action by preferences that are intrinsic to the agent, in a way that is not merely based on what we think is possible for that agent.

So, the causality of what you call "the odds that a god bothers" to do things in a particular way cannot be used by me as part of my model, since I can't imagine how to model the causality of a godlike psychology. Sorry, that level of causal modeling's out of my league.


William, if this argument narrowed the possible actions a god would take, you would have a point in your favor. Yet you're making the exact opposite claim. You're disclaiming the possibility of narrowing down what a god would want.

My claim is that, not knowing what a god/designer would do (in the generic case), we have to distribute our probability across all the possible things a god would do. The more things possible, the more life-bearing worlds possible, the better my argument works.

To defeat my argument, you have to make a predictive theory of God that shows that most worlds any god would create look like NDE worlds.

Heck, in fiction, most worlds we create aren't NDE worlds.

B. Prokop said...

But applying your analogy to the issue under discussion is completely bogus, due to multiple flaws. I'll just deal with two right now.

1. (For the sake of argument only - I do not actually buy into your going-in premise) given a 1% probability that Bob cooked the hamburgers, it is still possible for Bob to have done so. Your conclusion that evolution cannot have been guided on the basis of probability seems to ignore this possible outcome. After all, the huge odds against my winning the lottery do not preclude me from actually doing so.

2. There is no basis whatsoever to your going-in figures. I maintain and assert that you have manufactured figures that are favorable to the outcome you desire.

I myself am not going to take the trouble to manufacture competing figures, because I truly and deeply believe that this whole notion of "playing the odds" and discussing "Bayesian probabilities" when discussing philosophy or theology is total crap. It appears that contemporary philosophers have been suffering from "math envy" vis a vis the hard sciences, and the whole thing is a very mistaken attempt to paint a veneer of pseudoscience over an activity where such things have no genuine validity and do not belong. All these equations and math symbols are nothing more than a means to pretend that philosophy is somehow equivalent to empirical science, which it is not. (And that's not a bad thing!)

There. I've been wanting to say that for many months. I feel better now.

Crude said...

Some psychologies will. Some will not. So what? How does that help you? How does it shift the distribution of worlds that a god would create from plainly non-NDE worlds to plainly-NDE worlds?

Well, talking about 'plainly' is itself problematic, and opens up yet another problem. Does 'plainly' mean 'apparent'? And if so, apparent to who? According to what standard? We neither have a Gods-eye view of the world, nor is it the case that a world with NDE is so exclusively.

Beyond that, once you start talking about the psychologies of God, you're opening the door to all manner of vague speculation. Would Anselm's God create a universe that required periodic tune-ups? Does God value miniscule variations? What about non-life considerations - does God value life AND a consistent history? Does God value the creation of life that can comprehend its origins?

A lot of fiction involves designed or non-NDE worlds. Non-NDE worlds are more interesting to us because more is possible in them.

A tremendous amount of fiction that involves non-NDE worlds also is partially NDE - or those non-NDE worlds leave whether they are designed or not obcsure. Technically, *our* world is no longer an NDE-only world - we have Monsanto, and it's only going to get wilder from there, barring disaster. You don't want to talk 'interesting possibilities', because that's fundamentally technology talk.

Likewise, as I've pointed out, 'non-NDE world' != 'designed world' in principle. The spread of possibilities for just plain 'undesigned universes' is arguably more unbounded than designed universes - it includes just about anything you can imagine.

To defeat my argument, you have to make a predictive theory of God that shows that most worlds any god would create look like NDE worlds.

Not really. You have to accept that the principle spread of options for 'non-designed worlds' goes way, way beyond NDE. You seem to think you can ignore that, and compare the actual world to the huge spread of hypothetical design possibilities - when the proper comparison is between the spread of hypothetical design and non-design possibilities. And the possibilities in 'universe, not-designed' is freaking immense.

Doctor Logic said...

Bob,

My conclusion is that it is very unlikely that life was designed, not that it is impossible.

ozero91 said...

Did anyone actually read the paper? The thing is like 29 pages long. I have yet to do so.

And do we have any mathematicians here to interpret grod's comment?

ozero91 said...

This is what happens when you take a metaphysical concept and try to turn it into a "hypothesis" in an attempt to "explain the data."

Sammy Onomato said...

B. Prokop wrote:
“Because that's a philosophical/theological statement, and not a scientific one. I have zero objection to you or anyone else asserting that the processes are unguided, as long as you understand that.”

Stating that atomic decay or that biological evolution is an unguided process is no more a philosophical/theological statement than stating that there are guided and unguided rockets is a philosophical/theological statement.

You are using the word “guided” in a very specialized sense. Under you usage even things like unguided rockets would be guided by God. Or a rock rolling randomly down a hillside would be counted as a guided process. Or the roll of a die would be considered to be guided.

I have no problem with you using “guided” in this very special, theological sense. I do have a problem with your claim that scientists cannot make a scientific statement when using the word in its standard usage.




ozero91 said...

"I have no problem with you using “guided” in this very special, theological sense. I do have a problem with your claim that scientists cannot make a scientific statement when using the word in its standard usage."

Scientists practice Methodological Naturalism. When they describe a process, they assume, for the sake of convenience, that it is not guided. They leave the question of whether or not it is actually guided to philosophers. At least, that's how it would work out in a perfect world. But in the real-world you get people who extract metaphysics from the method. I suspect Bob's criticism was directed at the latter.

Sammy Onomato said...

ozero91,
But surely the fact that someone can use a scientific claim to support a particular metaphysical claim does not disqualify that claim as being scientific does it?

Crude said...

But surely the fact that someone can use a scientific claim to support a particular metaphysical claim does not disqualify that claim as being scientific does it?

If they're equivocating about what science shows or can show, then yes, it's disqualified. Not just as a scientific claim, but as a point whatsoever.

William said...

ozone:

I recall I read the paper the last time Vic posted it, and just skimmed it this time. Note that the paper says that DLogic's argument is theology (not just logic), since he is asserting knowledge of what his hypothetical God would choose:

"
But it might reasonably be pointed
out that, if God is one of the agents under consideration, any attempt to determine His motives
and abilities a priori is certainly an exercise in theology. This, in fact, is one of the reasons that a
Bayesian inference form appears disadvantageous to design arguments when the agents are not
human. For it seems possible that we will be required to say ahead of time what God (or even
aliens) would appreciate, value, and desire, and to describe the extent and limitations of their
abilities.
"

-- McGrew, Lydia, "Testability, Likelihoods, and Design." Philo 7:1 (Spring-Summer 2004), page 17.

Crude said...

ozero,

I've read through the paper. Still digesting it. The most interesting part is a part I've basically heard before - the vehicle in orbit around Alpha Centauri, and being able to make a design inference about it. If the response is we can't make an design inference (once humans are ruled out), then that's going to seem pretty absurd. If we could make a design inference in such a situation, then it's going to be asked why we can't make one in other areas, like biology.

Doctor Logic said...

ozero91,

I disagree with the statements about design being undetectable with science. I think that's nonsense.

I'm pretty sure there are animal experiments (certainly early childhood experiments) that test whether an agent is predicting the future and altering actions to bring a desired outcome to pass. That's design, folks.

Moreover, Paley is right. A watch on the beach is quite convincingly evidence of design. Even cogs of refined metal are signs of intelligent design because they have utility to human designers. Bacterial flagella would also be signs of design if they weren't found in the deep in the context of NDE, and if they had utility to some reasonably proximate designer.

What is design? Well, it's a situation wherein I can predict the outcome of my actions, I evaluate those outcomes with respect to my values and preferences, and then pick the action that most satisfies them. We could add criteria, like there must be a sequence of actions in order to call it design (so, say, deciding to eat a Cheeto instead of a potato chip is disqualified as design). But this is a scientifically testable situation.

Doctor Logic said...

Crude,

The most interesting part is a part I've basically heard before - the vehicle in orbit around Alpha Centauri, and being able to make a design inference about it. If the response is we can't make an design inference (once humans are ruled out), then that's going to seem pretty absurd. If we could make a design inference in such a situation, then it's going to be asked why we can't make one in other areas, like biology.

Yes!!!

But, I wonder, how do you square this with your earlier statements about designed-looking structures not being any less likely (in the Bayesian likelihood sense) with non-design?

Maybe there's just an as-yet undiscovered physical law about Volkswagens in orbits, or an as-yet undiscovered physical law of spontaneous generation of automobiles. If you find non-design implausible for the Volkswagen in orbit, then we should be in agreement.

Doctor Logic said...

William,

Note that the paper says that DLogic's argument is theology (not just logic), since he is asserting knowledge of what his hypothetical God would choose.

No. I'm asserting ignorance of what a hypothetical god would choose.

Suppose we're watching a horse race. We haven't seen the program, and know nothing about horse racing. Who will win the race?

If I a priori give equal odds to all the horses, am I asserting that I know which horse is going to win or that I know which will lose?

Clearly, that's exactly what I'm not doing. I'm asserting my ignorance about who will win. I don't know which horse will win, but if one wins, then the probability of victory by the competitors as a group adds up to 1.

Another example. Suppose there are two races. The 1pm race featured 99 horses, one of which was white with black feet.

The 2pm race featured 4 horses, all of them white with black feet.

Suppose all you know is that I bet and won on a white horse with black feet. You must conclude that there is a 99% chance that I bet and won on the 2pm race.

In making this inference, would you be asserting that you knew who won the 1pm race, or that you knew who was likely to win it?

Nope. Instead, you would be expressing your ignorance of who would win it, and merely expressing what you knew about who could win it!

ozero91 said...

"I disagree with the statements about design being undetectable with science. I think that's nonsense."

That's not what I meant to say, I was speaking to Sammy more along the lines of Aristotle's four causes in nature.

ozero91 said...

Substanstial form v.s. Artifacts, etc.

Crude said...

But, I wonder, how do you square this with your earlier statements about designed-looking structures not being any less likely (in the Bayesian likelihood sense) with non-design?

What I said was the following: when we're off in the land of pure hypothetical possibilities about designed v undesigned worlds, putting metaphysics/etc aside as much as we can, then just about any artifact you'd expect to find in the designed world is one you'd also expect to find in the undesigned world.

You really haven't had a response to this, and I think at the end of the day you're going to have to admit there isn't one. 'Undesigned worlds' permit all manner of crazy laws, brutely existing situations, etc, that go way, way beyond NDE. On the flipside, evolution is demonstrably a tool designers can use for ends.

Which means that, in terms of the Bayesian analysis you've been proposing ('What you'd expect in designed worlds v what you'd expect in non-designed worlds'), you really wouldn't be able to say the vehicle in question is designed.

Sure, that seems ridiculous, doesn't it? And yet, that's where you have to end up with your analysis when you're talking about what's in principle possible when treating our universe as a random selection between the spread of designed worlds versus the spread of undesigned worlds.

Bacterial flagella would also be signs of design if they weren't found in the deep in the context of NDE, and if they had utility to some reasonably proximate designer.

First, 'reasonably proximate'? What's 'reasonably proximate'? That sounds like a cop-out, since we both know that such a thing could in principle have all kinds of utility to a designer.

Second, whether it's 'deep in the context of NDE' is exactly what ID proponents question. Now, you can likely place it in an evolutionary context as opposed to specifically NDE - but, as Behe and Dembski would both say, there are all kinds of evolution. All are compatible with a designer (speaking scientifically), but some are more directly reliant than others.

Now, if you accept Paley, then the case ID is making is in principle one you have to deal with. I don't think it goes very far to insist 'No, NDE can accomplish this. I can't explain how this could be, much less how it could be at all reasonably to expect, but NDE predicts I wouldn't be able to say very much anyway!'

Crude said...

What is design? Well, it's a situation wherein I can predict the outcome of my actions, I evaluate those outcomes with respect to my values and preferences, and then pick the action that most satisfies them. We could add criteria, like there must be a sequence of actions in order to call it design (so, say, deciding to eat a Cheeto instead of a potato chip is disqualified as design). But this is a scientifically testable situation.

'Values' and 'preferences' aren't exactly scientific concepts. Further, being able to test some kinds of design wouldn't mean being able to test all kinds of design. That's exactly why you're having to come up with some kind of raw a priori analysis here.

Zach said...

DL wins this one, because the point he is making, once you cut through the verbiage on both sides, follows from basic arithmetic in trivial ways. The various attempts to dispute that have been mathematical confusions.

Zach said...

Crude please go take some basic probability/statistics, you just embarrass yourself.

Crude said...

DL wins this one, because the point he is making, once you cut through the verbiage on both sides, follows from basic arithmetic in trivial ways. The various attempts to dispute that have been mathematical confusions.

No, Zach - they haven't been.

DL's contention has been, from the start, that there's only 1 way life could exist in an undesigned universe, but a tremendous number of ways it could come to exist in a designed universe. This was incorrect. At this point, DL isn't even disputing that that's incorrect: for any conceivable 'created by a designer' universe he could imagine, you could imagine that same universe existing with the same results, sans designer. It will simply exist with the appropriate laws, or brute existences, or... etc.

Any universe you can imagine a designer creating, you can imagine 'just existing', if you're going into the hypothetical view with minimized metaphysics or priors. That's the point, and the problem for DL.

By all means, Zach - you say it's 'basic probability'. You'd love to show me wrong here, so here's your opportunity: argue your point. Show me where I'm wrong.

Keep in mind that, to do so, you're going to have to defend DL's original contention: there's only one hypothetically possible way for life and diversity to exist in a universe that is not designed.

You're not going to do this, because you can't. And you can't because you don't understand that the problem here isn't coming from the math side of things, but the hypotheticals. I never disputed the result of the calculations for the 99 recipes v 1 recipe example - I disputed whether saying one side only has 1 recipe is accurate.

But really, this isn't about probability, or reasoning, or much else for you. As usual, all you can do is register 'Crude saying X - I better say !X because me no like Crude!' So you blurt out whatever you can to disagree, and make a damn fool of yourself in the process.

I never tire of it. That's exactly why I'm giving you the opportunity to keep doing it, right now.

Bilbo said...

I tried reading all the comments and then realized that they go on almost ad infinitum. I think Dr. Logic is correct that a process that can be explained without design strengthens the hypothesis that the process wasn't designed. The more probable such an explanation of non-design is, the more probable that the non-design hypothesis becomes: e.g., NIST claims to have a computer model that explains why WTC7 was brought down by fires. The more probable that their computer model is, the more probable that WTC7 was brought down by fires. It would help if they released the data for their computer model, so that experts could determine how likely such a scenario is. So far, they refuse to do so, citing "public safety" issues. Of course, even if the computer model was probable, if it failed to explain other data,( such as nano-thermite in the WTC dust, or free-fall acceleration for 2 seconds, or eutectic corrosion of steel beams and columns, or continued ground temperatures well-above those of office fires for weeks after the collapse), then the overall probability of fires bringing down WTC7 may still be rather low.

Likewise, even if a theory for the origin of life and its evolution appeared that didn't require design appeared to be very probable, that still might not reduce the overall probability of the design hypothesis, based on other evidence. I disagree with Dr. Logic that Behe's case for irreducible complexity has been refuted, and there certainly hasn't been a plausible non-design scenario for the origin of life. But overall, I think Dr. Logic's logic is valid.

William said...
This comment has been removed by the author.
Doctor Logic said...

Crude,

Thank you for the stimulating discussion. I think you pushed a promising line of attack, and that got me to think about things in a lot more detail. I fact, I think the discussion got a few folks to think about the issue of inference, even if they disagree.

I'll just post my response and summary, and, if you're interested, we can continue on a new thread somewhere (my blog perhaps).

The general argument I've been making is that Bayesian reasoning rewards theories that make specific predictions over theories that make vague ones. NDE may seem broad and non-predictive to people who haven't thought about it, but NDE predicts descent, common descent, common composition and a few other things. NDE won't be knocked over by finding a second line of descent from an early form, but it could have been devastated had we found that cat and dog genomes were radically different, or if we discovered hedgehog factories buried underground, ghosts, etc. It's the great success of NDE (and organic chemistry) that makes us think spontaneous generation is very unlikely.

Given that specific, statistically-confirmed theories win Bayesian contests, my next move was to say that NDE beats any more vague theory, of which I considered design to be an example. How do we assess these sorts of inferences? Well, we imagine going back to having no data, outlining what we might expect if each theory were true.

If we knew there was life, but all we knew is that it was designed, we would not expect to see NDE take up more than the tiniest (infinitesimal is a better word) sliver of the probability space. Thus, design in the vague sense (i.e., without a detailed set of predictions about what we will find and why) is ruled out when we find NDE.

Your critique is that "non-design" is also a vague theory that would be ruled out in the same way. If it's true that any conceivable form of life could be expected under "non-design" as much as under "design", then my inference has little power to make inferences about whether life was designed. This is actually the best critique of my argument that I have ever seen.

Continued...

Doctor Logic said...

My initial reaction was to counter with a detailed specification of what constitutes a theory. Bayes theorem is about explanatory theories, and not merely restatements of observations, since P(O|O)=1. I suspect that the alleged multitude of non-design alternative life-bearing worlds evaporate when we restrict our considerations to actual theories rather than just brute facts. A non-design world that spontaneously generates protozoans doesn't explain protozoans anymore than a random production of protozoans from the mind of a god. Neither is explanatory. An explanatory design world is a world that represents a computed solution to a preference problem, not merely a random outcome implemented magically. We may be integrating over all possible preferences and solutions, but random outputs aren't design. Likewise, random non-design generations aren't explanatory theories either.

However, I think the much better response is to simply put everything that isn't physicalist reductionism on the other side of the fence. Instead of building the fence between design and non-design, I can put the division between mechanistic, physicalist, reductionist theories (MPR) on the one hand, and everything else on the other. The argument works in exactly the same way, and it shows why the confirmed predictions of evolutionary biology rule out spontaneous generation and design. If we do ignore my concerns about what constitutes a theory, and accept spontaneous generation as a theory, the majority of spontaneous generations won't remotely resemble the kinds of things that can be produced under reductionism.

While you might believe that the class of non-design worlds is as broad as the class of design worlds, I think you would agree that there are no other known MPR alternatives to NDE. Thus, my original inference stands.

Lastly, this same inference effectively excludes a god that designs an apparently physicalist world in order to create life. This is because, as in other cases, we're asking about whether an apparently physicalist NDE world is an MPR NDE world (mundane or even inescapable for MPR life-bearing worlds) or a world designed for life (whether MPR or not) which would be one of those extraordinary rare worlds designed to host life that happens to look like an MPR NDE worlds.

I anticipate objections based on my definition of "reductionist" and "physical". Feel free to experiment with redrawing the class distinctions. The point is that if the world has a physicalist basis, NDE is the only known mechanism for creating life. If life were not created with these non-design constraints, NDE is not expected. And, as you point out, if life is a brute fact of non-design outcomes, NDE isn't expected either.

William said...

DL: Your discussion of choosing a reductionist cosmos as the best for your argument puzzles me in one regard: even in a purely unguided reductionist scenario, for life to begin to evolve, don't we need spontaneous generation of life?

Doctor Logic said...

William,

Abiogenesis (which is required under reductionism) is not the same as spontaneous generation.

http://en.wikipedia.org/wiki/Abiogenesis

SG was intended to explain how maggots appeared in meat, and it typically refers to the rapid creation of complex life.

Abiogenesis under reductionism uses only the rules of chemistry, and is apparently quite rare.

William said...

DL: I fully understand that spontaneous generation was once used to explain the new growth of things like maggots on spoiled food when we merely could not see their source. (I recall I reproduced that historic experiment of Francisco Redi's with netting as a child in the 60's btw, and it smelled really bad).

Are you being pedantic to avoid the issue here? In most generic causal-based classifications, one would put abiogenesis and spontaneous generation in the same category.

So, pardon me,I'll rephrase below:


Your discussion of choosing a reductionist cosmos as the best for your argument puzzles me in one regard: even in a purely unguided reductionist scenario, for life to begin to evolve, don't we need [abiogenesis] of life?

BenYachov said...

>Abiogenesis (which is required under reductionism) is not the same as spontaneous generation.

Philosophically and metaphysically it clearly is.

Spontaneous Generation is primitive life coming from non-living matter via natural processes. All those who postulated Spontaneous Generation saw it as a phenomenon within nature not something from outside nature like a Jinn or "god" making it happen.

Now in terms of mechanism it was seen as a very fast process and Abiogenesis need not be so since it can take millions of years of evolution to transit from non-living to living.

But metaphysically & Philosophically Abiogenesis is clearly life coming from non-living mater via natural processes.

Again we are back to philosophy. It cannot be escaped from.

Doctor Logic said...

William,

Your discussion of choosing a reductionist cosmos as the best for your argument puzzles me in one regard: even in a purely unguided reductionist scenario, for life to begin to evolve, don't we need [abiogenesis] of life?

Yes, absolutely. I wasn't trying to say there was no need for abiogenesis. I assumed you were equating abiogenesis in a reductionist system with the non-reductive variety, because the no-reductive kind would conflict with my argument.

Basically, reductive abiogenesis is the kind of abiogenesis consistent with chemistry and statistical mechanics. This means that complex forms like amoebas won't just coalesce out of organic soup in a single step.

In a non-reductive abiogenesis, the kind of abiogenesis I associate with the term "spontaneous generation", an additional law emerges non-reductively (in addition to the laws of chemistry) to make amoebas coalesce spontaneously. Non-reductive abiogenesis is the biological equivalent of positing a law that Volkswagens or wristwatches will spontaneously appear wherever their raw ingredients can be found, and in violation of statistical mechanics.

William said...

If abiogenesis is rare and unlikely, then there is a bootstrapping issue to the reductive version of the universe, since we exist.

So I was wondering why choose the reductive version over the nonreductive one? Seems to me it makes the origins of life independent of design less likely?

Doctor Logic said...

Reductive abiogenesis is rare and unlikely, but not so rare as to be problematic. Exobiologists think it is likely to occur at least once within a few million years on a planet under conditions in the habitable zone.

By rare and unlikely, I meant that it's a factor in an established biosphere like ours. You don't have to worry about abiogenesis occurring in the sandwich your wife packed you for lunch. Abiogenic life, if it forms today, is too inefficient to compete with bacteria or is too fragile in the face of our oxygen atmosphere.

But even if reductive abiogenesis is rarer still (e.g., if reductive abiogenesis on our world was a freak accident), non-reductive abiogensis fares worse in our analysis because the non-reductive kind could do so much that we don't see. It's easy to imagine a non-reductive abiogenesis that breeds radically new complex species of bacteria or entire new forms of complex life in a continuous, hourly fashion. There are very many more ways of doing abiogenesis non-reductively than reductively, so finding ourselves in a world consistent with reductive abiogenesis implies that reductionism is extremely likely to be true.

Crude said...

NDE may seem broad and non-predictive to people who haven't thought about it, but NDE predicts descent, common descent, common composition and a few other things.

I don't think this is correct. Let's go down the list.

'Predicting descent' is utterly trivial. I think this you'd agree with.

'Common descent'? It barely predicts this so much as it's compatible with it: if there were multiple distinct lines of ancestry, NDE could absorb it. Nothing about NDE requires it in the broad sense, especially since NDE leaves OOL out of the question.

'Common composition' is vague.

So no, I think the amount of essential prediction built into NDE is minimal, at least with regards to the subject at hand. It's been retrofitted to an extreme already, and it will continue to be so.

Instead of building the fence between design and non-design, I can put the division between mechanistic, physicalist, reductionist theories (MPR) on the one hand, and everything else on the other.

First, make this move and you've expressly abandoned a scientific inference for a philosophical and metaphysical one - you're talking 'physicalist', 'mechanistic' and 'reductionist'. Philosophical and metaphysical concepts. That alone should give you pause.

Second, insofar as the conversation shifts towards the metaphysical and philosophical rather than the scientific, you have a compatibility issue on your hands - because various holistic, non-mechanistic, non-physicalist metaphysical and philosophical views are compatible with the science of NDE as is. Which among the selection is 'correct' is an open question.

Third, you're making this move at a bad time. Have you seen the responses to Nagel's recent book? Without getting into the contents of that book itself, the preferred line of defense against Nagel is to argue that reductionism isn't essential to NDE, and not only that, but that reductionism is a very unpopular idea, to the point where Nagel's attacking it is close to attacking a strawman. These responses come from self-described naturalistic philosophers.

Crude said...

While you might believe that the class of non-design worlds is as broad as the class of design worlds, I think you would agree that there are no other known MPR alternatives to NDE.

I would not agree, and my reasons for not agreeing would be similar to why I didn't agree with the design v non-design comparison. I've got other reasons why I think your line of argument isn't working (I just stated a few), but this one presents a fresh problem.

First: switching the comparison class between 'reductionist worlds' and 'non-reductionist worlds' doesn't matter, because that isn't your target to begin with: you're after designed v undesigned worlds and what we'd expect on either. Framing it specifically in terms of 'reductive physicalist' only complicates issues further for the reasons I already pointed out, and doesn't get you any further because the same problems that obtained with the reference to NDE are going to obtain with regards to 'reductive physicalist' insofar as what's at issue is design. You're back where you started.

Second, the difference between 'ruling out spontaneous generation' and 'ruling out design' is obvious: we have some scientific tests we can perform for the former, but not for the latter. If we want to see whether meat will spawn flies, we perform the experiment. If we want to see whether an omnipotent or even merely supremely powerful (think simulators in a Bostrom context) intelligent agent determined or determines the outcomes of evolution? There's no test. There's no control group. We're done.

Third, I don't think it's the case that there are no 'MPR alternatives to NDE' - in part, what qualifies as reductionist and physical are, again, wide open. Even the spontaneous generation case can be both - we can imagine that those things which have spontaneous generation don't themselves exist brutely, but come into existence via laws or processes which can be as complicated or simple as we like. Now, you may complain that eventually you're going to get to some brute facts, but here's your problem: that's true for the naturalist even in our world.

There are other problems, but for now I'll go with those.

Crude said...

Reductive abiogenesis is rare and unlikely, but not so rare as to be problematic. Exobiologists think it is likely to occur at least once within a few million years on a planet under conditions in the habitable zone.

Exobiologists have no idea about the likelihood of abiogenesis, period. They're pulling numbers out of the air. In fact, as of now, exobiology serves as a great example of a field without a subject.

Abiogenic life, if it forms today, is too inefficient to compete with bacteria or is too fragile in the face of our oxygen atmosphere.

How is this known? Again, what is this abiogenic life theory that doesn't amount to open-ended speculation?

But even if reductive abiogenesis is rarer still (e.g., if reductive abiogenesis on our world was a freak accident), non-reductive abiogensis fares worse in our analysis because the non-reductive kind could do so much that we don't see. It's easy to imagine a non-reductive abiogenesis that breeds radically new complex species of bacteria or entire new forms of complex life in a continuous, hourly fashion.

If we're limited only by our imaginations, all bets are off. This is doubly the case when you allow for 'reductive abiogenesis that occurs as a freak accident', because just what you can accomplish 'reductively, but with freak accidents' is absurd. To put this in perspetive: a boltzmann brain is a freak accident.

William said...

DL--It seems to me that you are adjusting the probability slider on abiogenesis to make life possible without new life forming in the sterile Petri dish. But until we survey a few thousand star systems for life,this is ad hoc and can't be used to argue probabilities versus design. Not enough data.

Crude said...

William,

It seems to me that you are adjusting the probability slider on abiogenesis to make life possible without new life forming in the sterile Petri dish. But until we survey a few thousand star systems for life,this is ad hoc and can't be used to argue probabilities versus design. Not enough data.

I think there's another possible confounding factor here.

What happens when/if intelligent agents (humans) start designing life?

I think that screws up these kinds of probability estimations even further beyond what I already said.

Edward T. Babinski said...

THEISM'S LACK OF REASON, Part 1

Why would it be necessary to create so many galaxies, stars, planets (over 200 billion known galaxies at last estimate), in order that a single planet might arise that could support life? Doesn't that seem like a lack of focus or a playing of the odds just a bit? Like the imagined Designer was flinging darts, loads of them, at a dart board the size of the cosmos?

Even then, life exists only on the quaking surface of one planet. Moving five miles above or beneath that thin surface, life dies. This is hardly the "filled" world/cosmos of the ancient Near Eastern imagination with the "world" filled with living things and shades beneath it, and God directly above it and not living light-years away. So modern theists make adjustments.

THEISM'S LACK OF REASON, Part 2

And why would it be necessary to "design" so many cousin species and cousins of cousin species, that all go extinct? Think of all the extinct species of lemurs and their cousin species and cousins of cousin species, and extinct species of ancient monkeys, ancient apes (including species of apes with larger cranial capacities than those of modern day apes), and extinct species of humans as well, that left behind much reduced numbers and species of lemurs, monkeys, apes, and humans. Does this shout "design?" Or again is it more like the dart board illustration I shared?

And what about universes beyond ours? For all we know there might also have been, based on analogies above, cousins of cousins of our own cosmos that appeared and became extinct before ours arose.









Edward T. Babinski said...
This comment has been removed by the author.
Edward T. Babinski said...

THEISM'S LACK OF REASON, Part 3

Here's a quotation from Chardin's writings that seems appropriate to mention:

So long as people believed, as St. Paul himself did, in one week of creation and a past of 4,000 years--so long as people thought the stars were satellites of the earth and that animals were there to serve man--there was no difficulty in believing that a single man could have ruined everything, and that another man had saved everything.

Pierre Teilhard de Chardin, “Fall, Redemption, and Geocentrism,”Christianity and Evolution
____________________________

I tend to agree with Chardin. Ancient Near easterners imagined that the gods created two major halves of creation, heaven and earth, and filled the earth with animals and humans, and the humans imagined the sky was filled with life too--with human-like divinities living directly overhead (not light-years away). They had no concept of our cosmos, so full of unused planets, barren real estate, so lacking in life and so filled with empty space, colliding galaxies, exploding stars, as well as new stars forming. So I agree with Chardin and with the folks below as well.
____________________________

Though it is not a direct article of the Christian faith that the planet we inhabit is the only inhabited one in the cosmos, yet it is so worked up from what is called the Mosaic account of creation, the story of Eve and the forbidden fruit, and the counterpart of that story, the death of the Son of God--that to believe otherwise renders the Christian system of faith at once little and ridiculous, and scatters it in the mind like feathers in the air.

Thomas Paine, The Age of Reason
____________________________

Did Jesus die uniquely to save the sins of human beings on planet Earth, or is he being strung up somewhere in the universe on every Friday?

Michael Ruse, “Booknotes,” Biology & Philosophy, Vol. 14, No. 1, Jan. 1999
____________________________

“A NEW HEAVEN?” EVEN FOR PEOPLE LIVING IN DISTANT GALAXIES?

According to the book of Revelation a “new earth” and a “new heaven” will be created after Jesus returns. Occupants of other planets throughout the more than hundred billion galaxies of our present “heaven” will no doubt be surprised to receive such an unearned favor, all because of what happens on our little planet. Or is this simply another example of how the Hebrews viewed the earth as the flat firm foundation of creation with the heavens above created simply for the earth below?
____________________________

SPEAKING OF I.D.ists none of them seem to be up on the conclusions of Ancient Near Eastern scholars found in the following online papers:

Mesopotamian Cosmic Geography in the Bible
http://biologos.org/uploads/projects/godawa_scholarly_paper_2.pdf

Biblical Creation and Storytelling: Cosmogony, Combat and Covenant
http://biologos.org/uploads/projects/godawa_scholarly_paper.pdf

Links to articles on the firmament by Paul Seely
http://theologicalscribbles.blogspot.com/2010/12/old-testament-cosmologypaul-seely.html

Genesis Through Ancient Eyes by John Walton
http://biologos.org/blog/series/genesis-through-ancient-eyes

And this recent piece that also leads to a video: "Interpreting Genesis 1, who's the literalist now?"
http://michaelsheiser.com/TheNakedBible/2012/09/interpreting-genesis-1-literalist/

Michael Heiser's view, per his piece above,is that "What Genesis describes is consistent with all other ancient Near Eastern creation models, and shares the vocabulary and motifs of those other pre-scientific cosmologies. Not a surprise, given God’s own choices about when to produce the material and who would do that."

THOUGH the point remains philosophically moot as to whether or not it was indeed "God" "choosing" to "produce" such "material."

Doctor Logic said...

Crude,

Frequentism?!

If we had to run an ensemble of planet-wide evolutionary scenarios before assessing the odds of the types of outcomes we could see, then ID would have no case. Who is to say the flagellum is unlikely to evolve without guidance? Have you done a statistically significant number of planet-sized simulations? We both know you haven't.

As a frequentist, you would have nothing to say about NDE except to say that ALL scientific generalizations (or even generalizations about suspects in criminal cases) are false because they infer rules that don't derive from frequentist statistics about the individuals involved.

But, of course, that's not what ID people do. It's not what any human does, for that matter. Any claim that NDE is inadequate implicitly assumes Bayesian reasoning. That means inferring probabilities from one-of-a-kind configurations. Lincoln was assassinated just once, but we're all pretty happy to cast judgment on what was likely to have occurred.

'Predicting descent' is utterly trivial. I think this you'd agree with.

What... I don't even... that's a ridiculous statement.

Statistically, you can't make a factory using NDE except by creating descent first, and then co-opting descent. For example, while mitochondria or bee hives might be considered factories of a sort, they can't just evolve from scratch by a series of small, statistically likely steps. In contrast, the non-reductive or design scenarios, an entire Ford Motor plant can spring up overnight.

As far as I am aware, humans have never invented any physical thing from scratch featuring descent. We've theorized about it, but it's generally something we don't want to do. We don't want automobiles breeding, feeding for themselves and consuming raw materials willy nilly.

When we use evolutionary computer algorithms to, say, build a better airfoil, we don't give the virtual airfoils reproductive organs, digestive tracts and hunting skills. There's no utility in that. Unguided evolution of simple life, on the other hand, requires integrated survival mechanisms and instincts. There are countless utilities a designer might have, but only one we can imagine for the early evolution of life (life before designers evolve).

Even if we humans did create beings that self-replicated (e.g., nanobots), we wouldn't use self-replication exclusively. Manufacturing has a lot of advantages in certain situations. We don't want self-replicating toasters.

Simply put, designers don't have the restriction of using descent to create simple life. NDE does.

Doctor Logic said...

'Common descent'? It barely predicts this so much as it's compatible with it: if there were multiple distinct lines of ancestry, NDE could absorb it. Nothing about NDE requires it in the broad sense, especially since NDE leaves OOL out of the question.

Flabbergasted. Really.

NDE does not allow for a new mammal to simply spring into existence at a late stage of evolution. Design does. Your magical form of spontaneous generation does. At best, NDE + chemistry could allow for new abiogenesis to occur and spawn a new line of protolife. Statistically speaking, that's it!

Creationists have long argued contra-NDE that species were poofed into existence by God in a fashion beyond the ability of NDE. Indeed, ID creationists do this when they whine about an evolutionary step being beyond NDE's capability. So I find your rejection of NDE's limitations to ring hollow.

'Common composition' is vague.

When humans made kevlar vests, this was an almost trivial modification from a nylon or metal plated vest. The only thing non-trivial about it is the kevlar itself. It's not as if no one had thought of making vests before kevlar came along. Vests have a simple, non-interacting design, which is held together by mechanical tension. Vest architecture is simple, and lacks feedback loops or spaghetti code, so substituting one raw material for another is trivial.

Evolved beings are built on millions of successive modifications to function, and you can't simply swap one compound for another unless you have a deep understanding of the machine's functional interrelationships. This understanding is something that NDE lacks. You can't replace bones with titanium rods without compensating in all the other biological systems at once. Something that biologists (and Behe) all agree upon. Still think this is vague?

NDE is far more restrictive than design. A designer can plan ahead, and modify the genome in thousands of places at once, and modifying thousands of interdependent functional units in order to create a radically different animal made of different stuff.

Doctor Logic said...

because various holistic, non-mechanistic, non-physicalist metaphysical and philosophical views are compatible with the science of NDE as is. Which among the selection is 'correct' is an open question.

Missing. The. Point.

This is like saying that rolling a 3 on a million-sided die is possible. True, but it doesn't have anything to do with interpreting a 3 as more likely to have been rolled on a 4-sided die than on the million-sided die. Doesn't affect my argument at all.

First: switching the comparison class between 'reductionist worlds' and 'non-reductionist worlds' doesn't matter, because that isn't your target to begin with

Make the interpretation of reductionism as narrow as it needs to be, and you'll see the argument work.

The narrow interpretation I'm using is as follows:

1) a world in which there is simple physics (i.e., folk mental concepts are not embedded into physics. No love in atoms, no awareness in quarks, just quantum fields),

2) a world where all structures form according to unplanned fluctuations in accord with statistical mechanics (there are no special laws that create life contra simple chemistry + statistical mechanics),

3) those bigger structures being simple chemical configurations of simpler substances.

This means no spontaneous generation of complex forms, only simple chemistry. H2O can readily form spontaneously, but bacteria will not form before their component molecules, nor will they form in violation of statistical mechanics of the simple molecules.

No VW's in orbit around Alpha Centauri.

As ID advocates love to write in their ridiculous screeds, the odds of a flagellum forming out of simple chemistry in solution is virtually zero. The odds of a tornado forming a 747 in a junk yard, are similarly almost zero. But wait! Stop the press! Crude says we shouldn't be surprised!

A world in which simple laws + statistical mechanics is violated in order to poof complex structures into existence is not in the space of worlds I'm considering.

This also means no design since there is nothing in simple physics to do any designing (until complex intelligent life has formed in the first place).

In other words, my in-space is like the standard physicalist picture in our universe. I think you know what that is.

The bulk of the complaints by ID proponents consists of claims that NDE alone is too restrictive to have created the life we observe with significant probability, whereas those probability constraints do not exist for designers. Here, for convenience to your defense, you're saying that these limitations are irrelevant to a probability analysis, since we can expect anything under reductionism and physicalism. It's hypocritical.

I'll add that the fine-tuning arguments thrown up by Christian apologists assume Bayesian reasoning, and also assume simple physics. So, again, I think you're quite familiar with the set of worlds I'm talking about.

Doctor Logic said...

Second, the difference between 'ruling out spontaneous generation' and 'ruling out design' is obvious: we have some scientific tests we can perform for the former, but not for the latter. If we want to see whether meat will spawn flies, we perform the experiment. If we want to see whether an omnipotent or even merely supremely powerful (think simulators in a Bostrom context) intelligent agent determined or determines the outcomes of evolution? There's no test. There's no control group. We're done.

And... now we're back to frequentism again. The whole point of Bayesian argument is to assess the probability of a theory being true based on the number of worlds consistent with each theory. Design is capable of much more than NDE + simple chemistry. NDE + spontaneous generation is capable of much more than NDE + simple chemistry. NDE + simple chemistry can only create life via NDE. We see NDE, so simple chemistry is the proper inference.

Cleave the space of possible worlds according to those in which NDE is the ONLY statistically expected solution for creating life, and infer from that space. You'll find that design and spontaneous generation of complex forms are effectively ruled out.

Doctor Logic said...

William,

DL--It seems to me that you are adjusting the probability slider on abiogenesis to make life possible without new life forming in the sterile Petri dish. But until we survey a few thousand star systems for life,this is ad hoc and can't be used to argue probabilities versus design. Not enough data.

Like Crude, you're trying to be frequentist about this. Frequentism is incorrect. Bayesian reasoning is correct.

A handful of laboratory experiments have shown that basic ingredients of life can form under conditions comparable with those on the early Earth, but no experiment has created protolife. This means abiogenesis is rather improbable.

We also know that abiogenesis does not occur on the friendly and stable conditions here on Earth, although that friendliness is severely limited by the presence of molecular oxygen.If abiogenesis occurred continuously, microbiologists and molecular biologists would have found it.

This establishes an upper bound on the frequency of abiogenesis in our region of the universe.

What about the lower bound?

Well, my inference shows why we're confident that abiogenesis occurred. If life were designed, abiogenesis did not have to start with single-celled life (as we know it did here).

Ever hear of the Garden of Eden? Newsflash: Adam and Eve are not single-celled life. God poofed them into existence. God doesn't have the limitations of NDE. He can do abiogenesis without being constrained by the statistics of simple chemistry.

If life is designed, we don't expect to see the fossil record that we do. Make a short list for yourself of the number of kinds of fossil records we might see if God designed a universe of life. What would the fossil record of Eden look like? What would the fossil record of Earth look like if it were 6000 years old? The fossil record of life on a designed planet would almost certainly look radically different.

Scientists believe abiogenesis happens frequently on a planetary scale because protolife formed on Earth very soon after the planet was cool enough to host it. Within 50 million years. This is fast on planetary timescales, and it's why scientists have started looking for simple life on other planets.