Get Real with Probability Assessments

Prompted by a TT blog comment and being curious, I looked for a blog entry alleged to show how blogger Chu-Carroll had demonstrated David Berlinski's cluelessness. I came upon Berlinski's Bad Math, a commentary on Berlinski's article On the Origins of Life. I came away disappointed for rather than showing bad math on the part of Berlinski it instead demonstrated Berlinski's failure to hold to faith based outcomes believed to have occurred in a storied prebiotic setting in which putative self-replicating molecules replicated their way to living cells through unknown and unidentified chemical pathways. From the link:

The odds of one self-replicating molecule developing out of a soup of pre-biotic chemicals is, according to Berlinksi, 1 in 10^60. But then, the replicator can't replicate unless it has a "template" to replicate against; and the odds of that are also, he claims, 1 in 10^60. Therefore, the probability of having both the replicator and the "template" is the product of the probabilities of either one, or 1 in 10^120.

Only that product formulation for combining probabilities only works if the two events are completely independent. They aren't. If you've got a soup of nucleotides and polymers, and you get a self-replicating polymer, it's in an environment where the "target template" is quite likely to occur. So the odds are not independent – and so you can't use the product rule.

Note the conditional assumptions. If you have a suitable soup and if you get a self-replicating polymer within this soup the target template is likely to occur. Not possessing the faith that either of these conditions would occur I wondered how one would assess the probability that they would. After all the entire scenario is contingent on the assumption that these conditions would have taken place. I would have expected one, who referred to his blog as goodmath, to at least make an effort at assessing the probability for requisite conditions before debunking set odds. Perhaps Berlinski's real error was in being too generous with his assessment of probabilities. But then a faith based story about putative self-replicators has no need of probability assessments.

This entry was posted on Wednesday, March 26th, 2008 at 10:49 pm and is filed under Origin of Life. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

59 Responses to “Get Real with Probability Assessments”

The Pixie Says:
March 27th, 2008 at 5:00 am
Note the conditional assumptions. If you have a suitable soup and if you get a self-replicating polymer within this soup the target template is likely to occur.

What do you think Chu-Carroll means by "quite likely" I suspect he means relative to 1 in 10^60 (and, I would say, dramatically more likely than that), rather than likely, i.e., p > 0.5. Big difference!

Not possessing the faith that either of these conditions would occur I wondered how one would assess the probability that they would.

Not sure I follow you. From what you have quoted, Chu-Carroll is not showing any "faith that either of these conditions would occur" either. Nor is he assessing the probability that they would as such.

After all the entire scenario is contingent on the assumption that these conditions would have taken place.

We need to be clear on the meaning of "condition" here. Chu-Carroll is talking about conditional probability, that is the probability of one event is different depending on whether another event occured or not. In this case Chu-Carroll is saying that the probability of the self-replicating molecule forming is much greater if the template has already formed than if the template has not. Of course that assumes the template formation has taken place, that is the point.

It does sound as though possibly you mean the physical conditions were just right, i.e., the temperature, the chemical mixture, etc.

I would have expected one, who referred to his blog as goodmath, to at least make an effort at assessing the probability for requisite conditions before debunking set odds.

I would guess Chu-Carroll believes we do not have enough information to be able to do that. Would you expect someone who was wanting to debunk evolution to have a full-fledged theory ready to take its place?

Perhaps Berlinski's real error was in being too generous with his assessment of probabilities.

Perhaps it was being too conservative. Chu-Carroll:

From this, he then argues that the odds of creating a self-replicating chain is 1 in 10^60.

Wow, that's a big number. He goes to some trouble to stress just what a whopping big number it is. Yes Dave, it's a big number. In fact, it's not just a big number, it's a bloody huge number. The shame of it is, it's wrong; and what's worse, he knows it's wrong. Right after he introduces it, he quotes a biochemist who pointed out the fact that that's stupid odds because there's probably more than one replicator in there. In fact, we can be pretty certain that there's more than one: we know lots of ways of modifying RNA/DNA chains that doesn't affect their ability to replicate. How many of those 10^60 cases self-replicate? We don't know. Berlinski just handwaves.

But then a faith based story about putative self-replicators has no need of probability assessments.

No, it needs laboratory work to test the chemistry and see what works. Which is what abiogenesis researchers are doing.

Comment by The Pixie — March 27, 2008 @ 5:00 am

Bradford Says:
March 27th, 2008 at 6:59 am
Pixie:

What do you think Chu-Carroll means by "quite likely" I suspect he means relative to 1 in 10^60 (and, I would say, dramatically more likely than that), rather than likely, i.e., p > 0.5. Big difference!

No. I think quite likely means what it appears to mean. IOW, that the likelihood of occurence was plausible. Plausibility in this case is contingent on the prior plausibility of enabling assumptions.

Not possessing the faith that either of these conditions would occur I wondered how one would assess the probability that they would.

Not sure I follow you. From what you have quoted, Chu-Carroll is not showing any "faith that either of these conditions would occur" either. Nor is he assessing the probability that they would as such.

Then he is making a scientifically vapid analysis. His whole approach incorporates an assumption of a self-replicator and a nourishing prebiotic soup. Their assumption can be granted as part of an if…then logical construction. But if empircal realities are considered as part of the strength and weakness of his analysis then his assessment of the probability that a self-replication scenario unfolds is what's the phrase?- yes that's it- vacuous.

After all the entire scenario is contingent on the assumption that these conditions would have taken place.

We need to be clear on the meaning of "condition" here. Chu-Carroll is talking about conditional probability, that is the probability of one event is different depending on whether another event occured or not. In this case Chu-Carroll is saying that the probability of the self-replicating molecule forming is much greater if the template has already formed than if the template has not. Of course that assumes the template formation has taken place, that is the point.

The larger context within which this skirmish occurs is a discussion of life's origins. Berlinski's essay addressed that larger point. The plausibility of template formation is very much relevant to the larger issue.

I would have expected one, who referred to his blog as goodmath, to at least make an effort at assessing the probability for requisite conditions before debunking set odds.

I would guess Chu-Carroll believes we do not have enough information to be able to do that. Would you expect someone who was wanting to debunk evolution to have a full-fledged theory ready to take its place?

Chu-Carroll's point needs to be analyzed for what it is. The belief that it renders Berlinski clueless is absurd.

Comment by Bradford — March 27, 2008 @ 6:59 am

Bradford Says:
March 27th, 2008 at 7:15 am
From this, he then argues that the odds of creating a self-replicating chain is 1 in 10^60.

Wow, that's a big number. He goes to some trouble to stress just what a whopping big number it is. Yes Dave, it's a big number. In fact, it's not just a big number, it's a bloody huge number. The shame of it is, it's wrong; and what's worse, he knows it's wrong. Right after he introduces it, he quotes a biochemist who pointed out the fact that that's stupid odds because there's probably more than one replicator in there. In fact, we can be pretty certain that there's more than one: we know lots of ways of modifying RNA/DNA chains that doesn't affect their ability to replicate. How many of those 10^60 cases self-replicate? We don't know. Berlinski just handwaves.

We can be pretty sure that there is more than one replicator? Arrogance. We don't even have the certainty there was one or the certainty of the existence of a soup loaded with pentose sugars, phosphate groups and nitrogenous bases. Berlinski was being polite when he waved his hand at you. Just saying hello.

Comment by Bradford — March 27, 2008 @ 7:15 am

agam Says:
March 27th, 2008 at 8:35 am
Bradford, now that you have taken the trouble to read through GoodMath, you should think of posting there. Chu-Carroll, doesn't need me to speak for him, even then I am going to make a couple of points on his behalf. Chu-Carroll questions Berlinski's use of probability to understand early earth chemical reactions (a scientifically better term than origin-of-life that is scientifically vacuous). And he points Berlinski back to the self-same article he quotes from. You know that thing about hammers and nails right? That's Berlinski. In the following exchange Berlinski had the grace to admit that he was making a heroic assumption about independent events. That was the end of the game for DB who hasn't been heard of since. So early chemical reaction conjectures contain assumptions? So what? Test them. Keep testing as many of them as you can. It is a very hard area, and may well remain a permanent unknown. That's the nature of science.

Nalam nalamariya aaval.

Comment by agam — March 27, 2008 @ 8:35 am

The Pixie Says:
March 27th, 2008 at 8:45 am
Bradford

No. I think quite likely means what it appears to mean. IOW, that the likelihood of occurence was plausible.

I will settle for plausible, that is fine with me.

Then he is making a scientifically vapid analysis. His whole approach incorporates an assumption of a self-replicator and a nourishing prebiotic soup.

Are you taking about Berlinski or Chu-Carroll? It is Berlinski who makes the assumptions first. It is Berlinski who thinks probability can be calculated. Does that make his analysis scientifically vapid?

The larger context within which this skirmish occurs is a discussion of life's origins. Berlinski's essay addressed that larger point. The plausibility of template formation is very much relevant to the larger issue.

I believe Chu-Carroll is a mathematian. Perhaps he feels qualified to comment on the maths and not the larger context.

Chu-Carroll's point needs to be analyzed for what it is. The belief that it renders Berlinski clueless is absurd.

Chu-Carroll's two basic points are (1) that the probability of a template forming is coditional on the replicatot forming and (2) that there could well be numerous repliocators possible. I do nopt see that you have addressed either. Indeed, I am not convinced you even understand the former.

We can be pretty sure that there is more than one replicator? Arrogance.

We can be sure there cannot? Arrogance.

Comment by The Pixie — March 27, 2008 @ 8:45 am

Bradford Says:
March 27th, 2008 at 8:59 am
agam:

Chu-Carroll questions Berlinski's use of probability to understand early earth chemical reactions (a scientifically better term than origin-of-life that is scientifically vacuous).

We are not lacking in knowledge as to the chemical nature of nucleic acids. Different sequence patterns correlate to differing biological functions. The sequence patterns are not explained by chemical bonds. What is vacuous is the contention that we are actually understanding early earth chemical reactions when what we are actually doing is making assumptions about them based on properties of biochemicals required by cells. No amount of phraseology can hide that.

In the following exchange Berlinski had the grace to admit that he was making a heroic assumption about independent events. That was the end of the game for DB who hasn't been heard of since.

Berlinski is smarter than his critics who do not have the grace to admit they are making not so heroic assumptions about conditions relevant to self-replicators in order to debunk an independent event scenario involving contingencies Berlinski is well aware of. DB is alive and well and not too impressed with the quality of the opposition.

Berlinski is all too well aware of the chemical properties of RNA. Here is another Berlinski quote:

Was nature lucky? It depends on the payoff and the odds. The payoff is clear: an ancestral form of RNA capable of replication. Without that payoff, there is no life, and obviously, at some point, the payoff paid off. The question is the odds.

For the moment, no one knows how precisely to compute those odds, if only because within the laboratory, no one has conducted an experiment leading to a self-replicating ribozyme. But the minimum length or "sequence" that is needed for a contemporary ribozyme to undertake what the distinguished geochemist Gustaf Arrhenius calls "demonstrated ligase activity" is known. It is roughly 100 nucleotides.

Comment by Bradford — March 27, 2008 @ 8:59 am

Bradford Says:
March 27th, 2008 at 9:06 am
Pixie: Chu-Carroll's two basic points are (1) that the probability of a template forming is coditional on the replicatot forming and (2) that there could well be numerous repliocators possible. I do nopt see that you have addressed either. Indeed, I am not convinced you even understand the former.

I understand that Chu-Carroll is posing a hypothetical to debunk a hypothetical. The hypothetical posed by him is weakly supported by empirical data. I'm not sure you understand that.

We can be pretty sure that there is more than one replicator? Arrogance.

We can be sure there cannot? Arrogance.

If this is about science and not ideology then it is incumbant to support SRM theories with empirical data. I don't have to be sure of anything. I only have to observe the lack of supporting evidence for SRM claims.

Comment by Bradford — March 27, 2008 @ 9:06 am

The Pixie Says:
March 27th, 2008 at 9:59 am
Bradford

I understand that Chu-Carroll is posing a hypothetical to debunk a hypothetical. The hypothetical posed by him is weakly supported by empirical data. I'm not sure you understand that.

Hmm, perhaps I do not understand. I thought Chu-Carroll was commenting on Berlinski's hypothetical. Perhaps you can clarify exactly what the difference is between the two hypotheticals.

If this is about science and not ideology then it is incumbant to support SRM theories with empirical data. I don't have to be sure of anything. I only have to observe the lack of supporting evidence for SRM claims.

Well this particular discussion is about maths. What in Chu-Carroll's argument do you think requires empirical data to support it? What does he have to be sure of? Nothing as far as I can see.

Comment by The Pixie — March 27, 2008 @ 9:59 am

Zachriel Says:
March 27th, 2008 at 10:00 am
Berlinski: The odds, then, are daunting; and when considered realistically, they are even worse than this already alarming account might suggest. The discovery of a single molecule with the power to initiate replication would hardly be sufficient to establish replication. What template would it replicate against? We need, in other words, at least two, causing the odds of their joint discovery to increase from 1 in 10^60 to 1 in 10^120.

Note the conditional assumptions. If you have a suitable soup and if you get a self-replicating polymer within this soup the target template is [not] likely to occur. Not possessing the faith that either of these conditions would[n't likely] occur I wondered how one would assess the probability that they would.

Comment by Zachriel — March 27, 2008 @ 10:00 am

Bradford Says:
March 27th, 2008 at 10:10 am
Pixie:

Well this particular discussion is about maths. What in Chu-Carroll's argument do you think requires empirical data to support it? What does he have to be sure of? Nothing as far as I can see.

Look harder.

Chu-Only that product formulation for combining probabilities only works if the two events are completely independent. They aren't. If you've got a soup of nucleotides and polymers, and you get a self-replicating polymer, it's in an environment where the "target template" is quite likely to occur.

Chu-Carroll's argument is predicated on the conditionals following the word if. The "target template" is quite likely to occur only if… In the essay written by Berlinski the plausibility of those conditionals is highly relevant to the main theme of his essay. Chu-Carroll does not have much of a case against the essay if he is confined to calculations based on his favored hypothetical.

Comment by Bradford — March 27, 2008 @ 10:10 am

The Pixie Says:
March 27th, 2008 at 10:25 am
Chu-Carroll's argument is predicated on the conditionals following the word if. The "target template" is quite likely to occur only if"¦ In the essay written by Berlinski the plausibility of those conditionals is highly relevant to the main theme of his essay. Chu-Carroll does not have much of a case against the essay if he is confined to calculations based on his favored hypothetical.

Chu-Carroll's argument says (using != for "not equal to"):

p(template|self-replicator) != p(template)

He justifies this by saying:

If you've got a soup of nucleotides and polymers, and you get a self-replicating polymer, it's in an environment where the "target template" is quite likely to occur.

Why do you think he is wrong?

Remember, you said:

After all the entire scenario is contingent on the assumption that these conditions would have taken place.

That would seem to support Chu-Carroll's argument that template formation and self-replicator formation are not independant events. If the conditions are right for one, then they are also right for the other.

Comment by The Pixie — March 27, 2008 @ 10:25 am

Bradford Says:
March 27th, 2008 at 10:35 am
If you've got a soup of nucleotides and polymers, and you get a self-replicating polymer, it's in an environment where the "target template" is quite likely to occur.

Pixie: Why do you think he is wrong?

Inconsequential is a better adjective. Conditionals are not logically wrong. Anyone can create conditionals. They are simply meaningless to the early earth chemistry agam alluded to.

Comment by Bradford — March 27, 2008 @ 10:35 am

One Brow Says:
March 27th, 2008 at 10:51 am
Then he is making a scientifically vapid analysis.

Chu-Carroll is not attempting a scientific analysis.

I would have expected one, who referred to his blog as goodmath, to at least make an effort at assessing the probability for requisite conditions before debunking set odds.

That would be your misunderstanding of how mathematics often works, then. You don't need to have an alternate calculation in hand to show a calculaiton is invalid. You don't need an alternate proof to invalidate a proof. All you need to do is show a flaw in the logic. Any one break in a chain of a mathematical proof or calculation invalidates the whole proof/calculation. This is the nature of a formal system.

Chu-Carroll's point needs to be analyzed for what it is. The belief that it renders Berlinski clueless is absurd.

The the Axiom of Choice implies Banach-Tarski paradox is absurd, yet it is true. Mathematicians work in the realm of the carefully structured absurd, and learn to play by its rules. Berlinski deviatied from the rules, and Chu-Carroll corrected him.

Comment by One Brow — March 27, 2008 @ 10:51 am

Doug Says:
March 27th, 2008 at 11:30 am
That would be your misunderstanding of how mathematics often works, then.

You and Zach sure like this type of line.

Comment by Doug — March 27, 2008 @ 11:30 am

The Pixie Says:
March 27th, 2008 at 11:58 am
Bradford

Inconsequential is a better adjective. Conditionals are not logically wrong. Anyone can create conditionals.

When someone says, after rolling fives sixes in a row, that the next roll of the dice cannot be another six, they are logically wrong. Or if they say it must be another six, they are logically wrong. The roll of a dice is not conditional on the previous rolls. Each roll is independent.

When Chu-Carroll says the formation of the template is conditional on the formation of the self-relicator, he is making the claim that p(template|self-replicator) != p(template). That claim might be wrong or it might be right. If he is right, then Berlinski's probability argument is wrong.

I do not understand how you can label that "Inconsequential", when you started a thread on that very issue!

They are simply meaningless to the early earth chemistry agam alluded to.

This thread is about Berlinski's probability calculation, and whether Chu-Carroll was right or wrong about two events being independent. What has that to do with agam, or with any "early earth chemistry" he may have alluded to?

Comment by The Pixie — March 27, 2008 @ 11:58 am

Bradford Says:
March 27th, 2008 at 12:23 pm
Pixie:

This thread is about Berlinski's probability calculation, and whether Chu-Carroll was right or wrong about two events being independent. What has that to do with agam, or with any "early earth chemistry" he may have alluded to?

If you retrace the comments on this and another thread you find that the Chu-Carroll blog was said to be evidence that Berlinski was clueless etc. Were it just agam I would chalk it up to his ignorance but apparently, based on web surfing, there are a number of like minded people. I thought Chu-Carroll's blog might be a slam dunk critique of Berlinski's essay. When I read it I realized it was merely evidence of closed minded thinking on the part of ID critics.

Comment by Bradford — March 27, 2008 @ 12:23 pm

Bradford Says:
March 27th, 2008 at 12:31 pm
OB: The the Axiom of Choice implies Banach-Tarski paradox is absurd, yet it is true. Mathematicians work in the realm of the carefully structured absurd, and learn to play by its rules. Berlinski deviatied from the rules, and Chu-Carroll corrected him.

Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

Comment by Bradford — March 27, 2008 @ 12:31 pm

One Brow Says:
March 27th, 2008 at 4:03 pm
You and Zach sure like this type of line.

You would prefer I don't mention a misunderstanding of the basic nature of what is needed for a mathematical proof?

Comment by One Brow — March 27, 2008 @ 4:03 pm

One Brow Says:
March 27th, 2008 at 4:06 pm
When Chu-Carroll says the formation of the template is conditional on the formation of the self-relicator, he is making the claim that p(template|self-replicator) != p(template). That claim might be wrong or it might be right. If he is right, then Berlinski's probability argument is wrong.

Actually, the criticism of Berlinski claims even less, stating that Berlinski needs to show p(template|self-replicator) = p(template). Chu-Carroll personally beleives p(template|self-replicator) > p(template), but that is not relevant to the criticism.

Comment by One Brow — March 27, 2008 @ 4:06 pm

One Brow Says:
March 27th, 2008 at 4:09 pm
Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

Chu-Carroll offered no correction at all. He pointed a basic flaw in Berlinski's work, and offered his (irrelevant) opinion on what the situation might be. However, regardless of Chu-Carroll's opinion, Berlinski's argumentative flaw is one I would not accept in a paper from a student in a 100-level statics class, and is all the more egregious from a mathematician.

Comment by One Brow — March 27, 2008 @ 4:09 pm

The Pixie Says:
March 27th, 2008 at 4:48 pm
Bradford

If you retrace the comments on this and another thread …

You started a thread focusing on one particular critism of Chu-Carroll. You seem to be abandoning that now. Do you now think that Chu-Carroll might have a point?

Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

Apparently not. So do you believe that p(template|self-replicator) = p(template)?

I ask this still unconvinced you actually understand what it means.

Comment by The Pixie — March 27, 2008 @ 4:48 pm

Bradford Says:
March 27th, 2008 at 5:02 pm
Pixie: You started a thread focusing on one particular critism of Chu-Carroll. You seem to be abandoning that now. Do you now think that Chu-Carroll might have a point?

He has no point worth considering.

Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

Apparently not. So do you believe that p(template|self-replicator) = p(template)?

I ask this still unconvinced you actually understand what it means.

I'm not convinced you know the difference between blind adherence to a belief and an analysis. Chu said "it's in an environment where the "target template" is quite likely to occur." That's an assertion, not evidence.

Comment by Bradford — March 27, 2008 @ 5:02 pm

Bradford Says:
March 27th, 2008 at 5:07 pm
Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

Chu-Carroll offered no correction at all.

LOL. Make up your mind. You had written: "…and Chu-Carroll corrected him."

He pointed a basic flaw in Berlinski's work, and offered his (irrelevant) opinion on what the situation might be. However, regardless of Chu-Carroll's opinion, Berlinski's argumentative flaw is one I would not accept in a paper from a student in a 100-level statics class, and is all the more egregious from a mathematician.

Why aren't the two events completely independent?

Comment by Bradford — March 27, 2008 @ 5:07 pm

The Pixie Says:
March 27th, 2008 at 6:39 pm
He has no point worth considering.

So do you think p(template|self-replicator) = p(template)? Why can you not answer that fundamental question?

Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

A fancy way of saying he is wrong. How is he wrong? Can you support that assertion? Or is it just wishful thinking?

I'm not convinced you know the difference between blind adherence to a belief and an analysis.

Well I did not start a thread about that. You started a thread about conditional probability (and still I find myself unconvinced you know that).

Chu said "it's in an environment where the "target template" is quite likely to occur." That's an assertion, not evidence.

It is an explanation of why he thinks Berlinski is wrong. As One Brow says, it is Berlinski making the claim, so the onus is on Berlinski to justify his implied claim that these two things are independent. All Chu-Carroll has to do is establish that one could be conditional on the other to make the calculation invalid.

Comment by The Pixie — March 27, 2008 @ 6:39 pm

Bradford Says:
March 27th, 2008 at 6:55 pm
So do you think p(template|self-replicator) = p(template)? Why can you not answer that fundamental question?

Pixie, where do you suppose nucleotides composing an imagined prebiotic replication come from?

Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

A fancy way of saying he is wrong. How is he wrong? Can you support that assertion? Or is it just wishful thinking?

Are you aware that the self-replication posited in theory is not observed in extra-cellular settings outside carefully controlled lab environments? I'm not the one wishfully thinking.

Chu said "it's in an environment where the "target template" is quite likely to occur." That's an assertion, not evidence.

It is an explanation of why he thinks Berlinski is wrong. As One Brow says, it is Berlinski making the claim, so the onus is on Berlinski to justify his implied claim that these two things are independent. All Chu-Carroll has to do is establish that one could be conditional on the other to make the calculation invalid.

What do you perceive to be the scientific significance of an argument over imagined scenarios never observed in nature?

Comment by Bradford — March 27, 2008 @ 6:55 pm

The Pixie Says:
March 27th, 2008 at 7:52 pm
Bradford

Pix: So do you think p(template|self-replicator) = p(template)? Why can you not answer that fundamental question?

Bradford: Pixie, where do you suppose nucleotides composing an imagined prebiotic replication come from?
Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

So do you think p(template|self-replicator) = p(template)?

Chu-Carroll does not say where the nucleotides come from, and his argument does not touch on that issue. Chu-Carroll is saying Berlinski is wrong to assert p(template|self-replicator) = p(template). You see completely unable to say whether p(template|self-replicator) = p(template) or not, and yet this is the fundamental issue of the thread you started. Making me more certain you do not understand what Chu-Carroll was saying.

Are you aware that the self-replication posited in theory is not observed in extra-cellular settings outside carefully controlled lab environments? I'm not the one wishfully thinking.

So we agree that your assertion was based on wishful thinking.

Chu said "it's in an environment where the "target template" is quite likely to occur." That's an assertion, not evidence.

What do you think the enironment is that he refers to here? I will tell you as I know you have a problem answering direct questions. The environment he is talking about is one in which a self-replicating forms. You may say such an environment is extremely unlikely; that is irrelevant to Chu-Carroll's point. He is saying Berlinski is wrong because if circumstances conspire to create an environment where a self-replicator forms, that self same environment will also be ideal for the formation of the template. Thus, the probability of template formation is conditional on self-replicator formation.

What do you perceive to be the scientific significance of an argument over imagined scenarios never observed in nature?

Oh, the argument is trivial. But imagined scenarios never observed in nature can certainly be important to science. Do you think we should reject any science we cannot observe in nature, for instance, an intelligent designer, perhaps, until we see the purported designer creating life?

Comment by The Pixie — March 27, 2008 @ 7:52 pm

Zachriel Says:
March 27th, 2008 at 8:04 pm
Bradford: What do you perceive to be the scientific significance of an argument over imagined scenarios never observed in nature?

That's precisely the point. Berlinski assigns probabilities to events that he has absolutely no idea how they may have occurred. And the probability he assigns to the second event *requires* that it be completely independent of the first event.

Comment by Zachriel — March 27, 2008 @ 8:04 pm

Bradford Says:
March 27th, 2008 at 8:08 pm
Pixie:

So we agree that your assertion was based on wishful thinking.

No, we're agreeing that your assertions are wishful thinking.:wink:

Chu said "it's in an environment where the "target template" is quite likely to occur." That's an assertion, not evidence.

What do you think the enironment is that he refers to here? I will tell you as I know you have a problem answering direct questions. The environment he is talking about is one in which a self-replicating forms. You may say such an environment is extremely unlikely; that is irrelevant to Chu-Carroll's point. He is saying Berlinski is wrong because if circumstances conspire to create an environment where a self-replicator forms, that self same environment will also be ideal for the formation of the template. Thus, the probability of template formation is conditional on self-replicator formation.

Pixie, you're missing a more fundamental argument made by Chu-Carroll. His argument is contingent on whether independent events occur. You and he have demonstrated only your personal preferences in that regard. There is no chemical rational supporting the lynchpin of Chu-Carroll's argument.

Comment by Bradford — March 27, 2008 @ 8:08 pm

Bradford Says:
March 27th, 2008 at 8:12 pm
Zachriel:

That's precisely the point. Berlinski assigns probabilities to events that he has absolutely no idea how they may have occurred. And the probability he assigns to the second event *requires* that it be completely independent of the first event.

Neither Berlinski nor Chu-Carroll has an idea as to how a non-occurence occurred. But what I want explained is what makes Chu-Carroll's assumption the scientifically preferred one?

Comment by Bradford — March 27, 2008 @ 8:12 pm

Zachriel Says:
March 27th, 2008 at 8:33 pm
Bradford: Neither Berlinski nor Chu-Carroll has an idea as to how a non-occurence occurred.

We're in agreement then. Berlinski multiplies nonsense times nonsense to match his predetermined conclusion.

Bradford: But what I want explained is what makes Chu-Carroll's assumption the scientifically preferred one?

Chu-Carroll's point is that the two Berlinski nonsenses are only equal if they are completely independent, and that Berlinski nonsense squared still doesn't make for a valid argument.

Comment by Zachriel — March 27, 2008 @ 8:33 pm

Bradford Says:
March 27th, 2008 at 8:37 pm
Zachriel:

We're in agreement then. Berlinski multiplies nonsense times nonsense to match his predetermined conclusion.

Naw. Berlinski's calculations are based on one physical assumption and Chu-Carroll's on the opposite assumption.

Bradford: But what I want explained is what makes Chu-Carroll's assumption the scientifically preferred one?

Chu-Carroll's point is that the two Berlinski nonsenses are only equal if they are completely independent, and that Berlinski nonsense squared still doesn't make for a valid argument.

Wrong. Chu-Carroll's point is that his calculation is correct and Berlinski's erroneous. So what makes Chu-Carroll's assumption the scientifically preferred one?

Comment by Bradford — March 27, 2008 @ 8:37 pm

Zachriel Says:
March 27th, 2008 at 8:57 pm
Bradford: Berlinski's calculations are based on one physical assumption and Chu-Carroll's on the opposite assumption.

Berlinski made many physical assumptions. That DNA was required. That each monomer to be bound in a sequence is independent of every other monomer. That the sequence of first replicator and the sequence of the first template were independent of one another. Chu-Carroll attacks the latter assumption.

Whatever the environment such a first replicator occurred in may have also encouraged the formation of suitable templates. Indeed, the assumption of a distinct replicator and template in a collection of interacting molecules may be faulty. If the replicator and template were the same molecular network, then the probability p(template|self-replicator) could be equal to one.

Comment by Zachriel — March 27, 2008 @ 8:57 pm

Bradford Says:
March 27th, 2008 at 9:21 pm
Zachriel:

Whatever the environment such a first replicator occurred in may have also encouraged the formation of suitable templates.

You can assume anything including the generation of a self-replicating RNA molecule but why would this be a likely event? These scenarios are so detached from physical realities that they are impossible to assess.

Indeed, the assumption of a distinct replicator and template in a collection of interacting molecules may be faulty. If the replicator and template were the same molecular network, then the probability p(template|self-replicator) could be equal to one.

I know how to weave if…then scenarios together too as does Berlinski. What I do not see from the Chu-Carroll camp are reasons grounded in chemistry by which we should view an independent event assumption that exclusively supports Chu-Carroll's POV.

Comment by Bradford — March 27, 2008 @ 9:21 pm

Zachriel Says:
March 27th, 2008 at 10:13 pm
Bradford: These scenarios are so detached from physical realities that they are impossible to assess.

Then we agree. Berlinski's calculations are based on faulty assumptions.

Bradford: I know how to weave if"¦then scenarios together too as does Berlinski.

The difference is that Chu-Carroll is more reserved in his assumptions.

Chu-Carroll: Only that product formulation for combining probabilities only works if the two events are completely independent. They aren't. If you've got a soup of nucleotides and polymers, and you get a self-replicating polymer, it's in an environment where the "target template" is quite likely to occur. So the odds are not independent – and so you can't use the product rule.

In any reasonable scenario, molecules will be interacting in complex manners, rendering any assertion of independence unsupportable. Indeed, independence would be the special case, and represents an additional assumption.

And then Berlinski makes this error.

Chu-Carroll: Not to mention that he repeats the same error he made before: assuming that there's exactly one "template" molecule that can be used for replication.

Which make Berlinski's calculations nonsense squared.

Comment by Zachriel — March 27, 2008 @ 10:13 pm

Bradford Says:
March 27th, 2008 at 10:51 pm
Zachriel: Then we agree. Berlinski's calculations are based on faulty assumptions.

His calculations are based on possible assumptions.

Bradford: I know how to weave if"¦then scenarios together too as does Berlinski.

The difference is that Chu-Carroll is more reserved in his assumptions.

Chu-Carroll: Only that product formulation for combining probabilities only works if the two events are completely independent. They aren't. If you've got a soup of nucleotides and polymers, and you get a self-replicating polymer, it's in an environment where the "target template" is quite likely to occur. So the odds are not independent – and so you can't use the product rule.

I disagree that target template is a quite likely occurence.

In any reasonable scenario, molecules will be interacting in complex manners, rendering any assertion of independence unsupportable. Indeed, independence would be the special case, and represents an additional assumption.

Why would a complex interaction of different molecules indicate a likelihood of a target template?

And then Berlinski makes this error.

Chu-Carroll: Not to mention that he repeats the same error he made before: assuming that there's exactly one "template" molecule that can be used for replication.

Which make Berlinski's calculations nonsense squared.

The problem is a very small subset of template molecules compared to an enormous number of disruptive reactions which are much more probable.

Comment by Bradford — March 27, 2008 @ 10:51 pm

The Pixie Says:
March 28th, 2008 at 5:48 am
Bradford

Brad: I'm not the one wishfully thinking.

Pix: So we agree that your assertion was based on wishful thinking.

Brad: No, we're agreeing that your assertions are wishful thinking.

No, we do not agree with that. You said we were both wishful thinking, I said you were wishful thinking. That means we agree you were wishful thinking, and only you think I was wishful thinking.

Pixie, you're missing a more fundamental argument made by Chu-Carroll. His argument is contingent on whether independent events occur.

No, you are not understanding his argument. His argument is based on mathematics of probability.

p(event A AND event B) = p(event A) x p(event B)

… if and only if event A and event B are independant. Chu-Carroll is pointing out that event A (formation of the self-replicator) and event B (formation of the template) may well not be independant, so Berlinski's maths is wrong. The actual probability of the events is not the issue (in the part you quote in the OP).

There is no chemical rational supporting the lynchpin of Chu-Carroll's argument.

That is because Chu-Carroll's argument is mathematical. You have not got that yet, have you?

Try this analogy for size. Dr B. is looking at eggs in a nest. On a certain farm, he estimates that the chances of a particular egg being a duck's egg is 0.3 {p(duck egg) = 0.3}. He comes across a nest with two eggs, and decides, therefore, that the probability of them both being duck eggs is 0.3 x 0.3 = 0.09.

Dr. C-C. disputes that calculation, saying Dr B.'s calculation is only valid if the identity of the second egg is independent of the identity of the first egg. He points out that if the first egg is a duck egg, then this is presumably a duck's nest, and all the eggs will be duck eggs. If the first egg is not a duck egg, then we would not expect any other of the other eggs to be duck eggs. The probability for the second egg is conditional on what the first egg is.

p(second duck egg|first duck egg) = 1 p(second duck egg|first not duck egg) = 0

This would mean that the probability of both being duck eggs is 0.3 x 1.0 = 0.3. Note that Dr C-C is not disputing the value of 0.3; Dr C-C. is a mathematician and recognises he is not in a position to talk with any authority on what the actual probability is, but he can point out the mathematical flaw in Dr B.'s argument.

Comment by The Pixie — March 28, 2008 @ 5:48 am

Bradford Says:
March 28th, 2008 at 7:03 am
That is because Chu-Carroll's argument is mathematical. You have not got that yet, have you?

Understand this well Pixie. Using mathematical symbols means zilch when attempts to quantify bear no resemblance to physical realities. We have yet to see an egg in the prebiotic soup and worse yet you and the others are blind to the condor hovering over the nest. But you're very impressed when a mathematician who knows nothing about chemical realities claims a valid mathematical representation of what is occuring.

Comment by Bradford — March 28, 2008 @ 7:03 am

The Pixie Says:
March 28th, 2008 at 7:32 am
Understand this well Pixie. Using mathematical symbols means zilch when attempts to quantify bear no resemblance to physical realities.

You, I and I suspect Chu-Carroll agree there. We just need to persuade Berlinski.

We have yet to see an egg in the prebiotic soup and worse yet you and the others are blind to the condor hovering over the nest. But you're very impressed when a mathematician who knows nothing about chemical realities claims a valid mathematical representation of what is occuring.

On the contrary, it is you who seems to be impressed when a mathematician (Berlinski) claims a valid mathematical representation of what is occuring. Earlier you were complaining that Chu-Carroll had not offered an alternative mathematical representation of what is occuring (remember this: "I would have expected one, who referred to his blog as goodmath, to at least make an effort at assessing the probability for requisite conditions before debunking set odds"), and now you seem to think I am very impressed with the assessment you said he did not make. How bizarre!

So that now just leaves us wondering where you stand on the discussion of conditional probability – the part of Chu-Carroll's argument you quoted in the OP ("Only that product formulation for combining probabilities only works if the two events are completely independent."). Are you prepared to say whether you think p(template|self-replicator) = p(template). If you are still having trouble understanding conditional probability, just say, and I will try to explain it more simply. Or look at the Wiki entry I linked to before.

Comment by The Pixie — March 28, 2008 @ 7:32 am

Zachriel Says:
March 28th, 2008 at 8:06 am
Bradford: Why would a complex interaction of different molecules indicate a likelihood of a target template?

Berlinski pulls two probability assessments out of his hat: p(D1) = 10^-60 and p(D2) = 10^-60, specific, and as you yourself have indicated, unsupportable. The only probability assessment Chu-Carroll has made is that p(D2|D1) "¹ p(D1). Please read The Pixie's "Duck Egg" explanation again.

â‰

Comment by Zachriel — March 28, 2008 @ 8:06 am

Bradford Says:
March 28th, 2008 at 8:32 am
Pixie:

On the contrary, it is you who seems to be impressed when a mathematician (Berlinski) claims a valid mathematical representation of what is occuring.

I never claimed to be impressed with his odds (although the essay on the whole is good) simply because neither he nor Chu is in a position to assesss them. Chu compounded the problem by pretending otherwise.

So that now just leaves us wondering where you stand on the discussion of conditional probability – the part of Chu-Carroll's argument you quoted in the OP ("Only that product formulation for combining probabilities only works if the two events are completely independent."). Are you prepared to say whether you think p(template|self-replicator) = p(template). If you are still having trouble understanding conditional probability, just say, and I will try to explain it more simply. Or look at the Wiki entry I linked to before.

If a demon exists on Mars does that increase the likelihood that a second demon will be found?

Comment by Bradford — March 28, 2008 @ 8:32 am

Bradford Says:
March 28th, 2008 at 8:34 am
Zachriel:

Berlinski pulls two probability assessments out of his hat: p(D1) = 10^-60 and p(D2) = 10^-60, specific, and as you yourself have indicated, unsupportable. The only probability assessment Chu-Carroll has made is that p(D2|D1) "¹ p(D1). Please read The Pixie's "Duck Egg" explanation again.

And based on this what do you think of the chances of finding a demon on Mars?

Comment by Bradford — March 28, 2008 @ 8:34 am

The Pixie Says:
March 28th, 2008 at 9:51 am
Bradford

I never claimed to be impressed with his odds (although the essay on the whole is good) simply because neither he nor Chu is in a position to assesss them. Chu compounded the problem by pretending otherwise.

Wait a minute. You know full well that Chu-Carroll never assessed the odds. You said earlier that you were surprised he had not: "I would have expected one, who referred to his blog as goodmath, to at least make an effort at assessing the probability for requisite conditions before debunking set odds." What a tangled web you weave, Bradford.

Now let us be quite honest here. Berlinski assessed the odds. You, I and Chu-Carroll agree he was in no position to do so. Chu-Carroll did not assess the odds. It would be dishonest to try to pretend that he did.

If a demon exists on Mars does that increase the likelihood that a second demon will be found?

Yes. Here is another one. If one day we discover there really is a monster in Loch Ness, it will dramatically raise the odds of finding a second one.

And based on this what do you think of the chances of finding a demon on Mars?

We have no reasonable way of assessing the probability. But we can still say that if we find one, the chances of finding another will be much higher. Think about this carefully; this is very close to Chu-Carroll's claim.

Comment by The Pixie — March 28, 2008 @ 9:51 am

Bradford Says:
March 28th, 2008 at 10:07 am
Now let us be quite honest here. Berlinski assessed the odds. You, I and Chu-Carroll agree he was in no position to do so. Chu-Carroll did not assess the odds. It would be dishonest to try to pretend that he did.

WB: If a demon exists on Mars does that increase the likelihood that a second demon will be found?

Yes. Here is another one. If one day we discover there really is a monster in Loch Ness, it will dramatically raise the odds of finding a second one.

Berlinski assessed odds for an initial self-replicator at 1 in 10^60. He would have been better advised to either stop there or simply state the truth with no fanfare. The simple truth is there is no scientific basis for assuming an initial self-replicator would arise. He allowed others to ignore a good essay in favor of attacking him and one of the paragraphs in the essay. Chu-Carroll did not lay out specific odds but he did make this remark which I have labored in vain to get others to address:

They aren't. If you've got a soup of nucleotides and polymers, and you get a self-replicating polymer, it's in an environment where the "target template" is quite likely to occur.

It was Chu-Carroll who introduced a chemically based justification for what followed. The trouble is he is wrong and neither he nor his followers have engaged in substantive discussions of his hopelessly vague pronouccement. Chu-Carroll puts forth an innocuous demon argument. It could have been candy bars on planet Zennoch in the Zudu galaxy for all the scientific import it has which is zero.

Comment by Bradford — March 28, 2008 @ 10:07 am

Zachriel Says:
March 28th, 2008 at 10:28 am
Chu-Carroll: Only that product formulation for combining probabilities only works if the two events are completely independent. They aren't.

Bradford: It was Chu-Carroll who introduced a chemically based justification for what followed.

The argument is that p(D2|D1) "¹ p(D1), per this:

Bradford: If a demon exists on Mars does that increase the likelihood that a second demon will be found?

Yes. Just as the probability of finding diamonds in a mountain are higher, given the first. Just as the probability of a second candy bar on Zennoch in the Zudu galaxy is higher, given the first.

Comment by Zachriel — March 28, 2008 @ 10:28 am

Bradford Says:
March 28th, 2008 at 10:36 am
Zachriel:

Yes. Just as the probability of finding diamonds in a mountain are higher, given the first. Just as the probability of a second candy bar on Zennoch in the Zudu galaxy is higher, given the first.

That succintly summarizes the state of abiogenesis. Increased probabilities based on assumed fantasies.

Comment by Bradford — March 28, 2008 @ 10:36 am

Zachriel Says:
March 28th, 2008 at 10:55 am
Bradford: That succintly summarizes the state of abiogenesis. Increased probabilities based on assumed fantasies.

*You* are the one who posited finding a demon on Mars.

Comment by Zachriel — March 28, 2008 @ 10:55 am

Bradford Says:
March 28th, 2008 at 10:58 am
Bradford: That succintly summarizes the state of abiogenesis. Increased probabilities based on assumed fantasies.

Zachriel: *You* are the one who posited finding a demon on Mars.

Yeah, but I'm not the one positing a self-replicator in a prebiotic soup and a cell at the end of that story.

Comment by Bradford — March 28, 2008 @ 10:58 am

The Pixie Says:
March 28th, 2008 at 11:06 am
Berlinski assessed odds for an initial self-replicator at 1 in 10^60. He would have been better advised to either stop there or simply state the truth with no fanfare. The simple truth is there is no scientific basis for assuming an initial self-replicator would arise.

Ah, so when Bverlinski assesses the odds, you are happy with that, but when Chu-Carroll does – or does not – he has "compounded the problem.". Do you think Berlinski's assessment is valid (i.e., does he base it on reasonable science)?

He allowed others to ignore a good essay in favor of attacking him and one of the paragraphs in the essay.

Boo and hiss to people who pick out specific paragraphs in an essay to highlight potential problems. Er, like you did in the OP.

Chu-Carroll did not lay out specific odds …

No, he did not. He knows as well as you and I that any such assessment would be nonsense. Oh, wait, you think it is not nonsense when Berlinski does it, only when Chu-Carroll does it.

Chu-Carroll did not lay out specific odds but he did make this remark which I have labored in vain to get others to address:
"They aren't. If you've got a soup of nucleotides and polymers, and you get a self-replicating polymer, it's in an environment where the "target template" is quite likely to occur. "
It was Chu-Carroll who introduced a chemically based justification for what followed.

Good lord no! Why did he not pluck figures out of thin air like Berlinski? The cad, using a "chemically based justification". What was he thinking?

So to summarise, you think what Berlinski did was good because he came up with a specific figure, even though we agree we do not have the information to do that, and did not have any chemically based justification. On the other hand, Chu-Carroll was wrong, because he did not give a specific figure, and he used a "chemically based justification".

Berlinski: Specific figure for probability, no chemically based justification
Chu-Carroll: No specific figure for probability, but chemically based justification

Think about this very carefully and see if you can see the irony inherent in your position. You might like to ponder whether a chemically based justification is a good thing or a bad thing. You might wonder whether it is reasonable to come up with a specific figure or not. Then perhaps you could address why you accept Berlinski and reject Chu-Carroll and consider whether it is due to their arguments or rather whether it is because of their ideaologies alone.

The trouble is he is wrong and neither he nor his followers have engaged in substantive discussions of his hopelessly vague pronouccement. Chu-Carroll puts forth an innocuous demon argument. It could have been candy bars on planet Zennoch in the Zudu galaxy for all the scientific import it has which is zero.

It is sad you cannot bring yourself to say the same about Berlinski. Berlinski is the one who contrived to find an actual figure for the probability, and yet you seem happy to accept that on faith. Chu-Carroll disagrees with you. Thus, people on his side are labelled "followers" (to indicate some kind of religion perhaps, or sheep following the pack). And clearly Chu-Carroll is wrong because, while he never gives a specific figure, that figure must still be wrong because he is on the side of the evil atheists.

Chu-Carroll's claim – the one you quote in the OP – is not vague. It is very precise, valid and relevant. The trouble is that you do not understand it and, I suspect, will not try to understand it for fear that he might be right.

If you can say whether p(template|self-replicator) = p(template) or not, and can offer some justification for your position, we will be able to engage in substantive discussions.

Comment by The Pixie — March 28, 2008 @ 11:06 am

Bradford Says:
March 28th, 2008 at 11:25 am
Pixie:

Ah, so when Bverlinski assesses the odds, you are happy with that, but when Chu-Carroll does – or does not – he has "compounded the problem.". Do you think Berlinski's assessment is valid (i.e., does he base it on reasonable science)?

Why are you ignoring this prior comment I made?

Pix: On the contrary, it is you who seems to be impressed when a mathematician (Berlinski) claims a valid mathematical representation of what is occuring.

Brad: I never claimed to be impressed with his odds (although the essay on the whole is good)

It is not the specific odds I concur with rather the indication that the arsing of a self-replicator is implausible as explained in the ignored part of Berlinski's essay.

So to summarise, you think what Berlinski did was good because he came up with a specific figure, even though we agree we do not have the information to do that, and did not have any chemically based justification. On the other hand, Chu-Carroll was wrong, because he did not give a specific figure, and he used a "chemically based justification".

Why do you omit key phrases like "hopelessly vague" which I used to describe the "justification." I'll spell it out for you this time. The "justification" is fatally flawed.

The trouble is he is wrong and neither he nor his followers have engaged in substantive discussions of his hopelessly vague pronouccement. Chu-Carroll puts forth an innocuous demon argument. It could have been candy bars on planet Zennoch in the Zudu galaxy for all the scientific import it has which is zero.

It is sad you cannot bring yourself to say the same about Berlinski. Berlinski is the one who contrived to find an actual figure for the probability, and yet you seem happy to accept that on faith.

Berlinski is correct in his argument and superfluous with his math. That's a scientific assessment not an expression of faith. The misplaced faith lies with Chu-Carroll syncophants who think his demon on Mars arguments are substantive.

Chu-Carroll's claim – the one you quote in the OP – is not vague. It is very precise, valid and relevant. The trouble is that you do not understand it and, I suspect, will not try to understand it for fear that he might be right.

I understand that there is an increased likelihood of a second demon on Mars if we find the first one. That's the argument for abiogenesis. A bogus science but a perfect religion for atheism.

Comment by Bradford — March 28, 2008 @ 11:25 am

The Pixie Says:
March 28th, 2008 at 11:54 am
Bradford

Why are you ignoring this prior comment I made?
"I never claimed to be impressed with his odds (although the essay on the whole is good)"

Because it is vacuous. If you are prepared to state that his calcuation of the odds is not reasonable, then we could move on in agreement. If you had said that you were not impressed with his calculation, I might have queried it, asking you to say one way or the other. But all you are saying here is that you never claimed to be impressed. So for all I know you might have been impressed with the calculation, and merely not said so.

It is not the specific odds I concur with rather the indication that the arsing of a self-replicator is implausible as explained in the ignored part of Berlinski's essay.

You started the thread focusing on a specific part of Berlinski's argument. Please do not complain if we subsequently focus on that specific part of Berlinski's argument!

Why do you omit key phrases like "hopelessly vague" which I used to describe the "justification." I'll spell it out for you this time. The "justification" is fatally flawed.

So far all we have is your assertion that this is the case. As Chu-Carroll's argument revolves around the rejection of p(template|self-replicator) = p(template), and you have been unwilling to say whether you agree with that or not, I am going to reject your assertion.

Berlinski is correct in his argument and superfluous with his math. That's a scientific assessment not an expression of faith.

Berlinski's argument is based on the maths. If the maths is wrong, the argument fails. That is why he put the maths in. Hmm, perhaps you think he put the maths in to confuse people? Or to make it seem more technical? Or to scare people away so they do not actually read the rest of it?

I understand that there is an increased likelihood of a second demon on Mars if we find the first one.

Excellent. Now can see how why this is wrong?

p(two demons on Mars) = p(demon on Mars) x p(demon on Mars)

I am guessing no, but hopefully you will prove me wrong on this one.

Comment by The Pixie — March 28, 2008 @ 11:54 am

Bradford Says:
March 28th, 2008 at 12:05 pm
Berlinski's argument is based on the maths. If the maths is wrong, the argument fails.

Wrong. The only failing would be the math itself. The argument that OOL is bogus is on firm ground. Have you read the entire essay?

I understand that there is an increased likelihood of a second demon on Mars if we find the first one.

Pixie: Excellent. Now can see how why this is wrong?

Have you changed your mind? You indicated previously that "if we find one, the chances of finding another will be much higher."

And based on this what do you think of the chances of finding a demon on Mars?

We have no reasonable way of assessing the probability. But we can still say that if we find one, the chances of finding another will be much higher.

Comment by Bradford — March 28, 2008 @ 12:05 pm

One Brow Says:
March 28th, 2008 at 12:12 pm
Bradford(1): Chu-Carroll's "correction" is predicated upon an erroneous view of biochemical prebiotic possibilities.

One Brow(1): Chu-Carroll offered no correction at all.

Bradford(2): LOL. Make up your mind. You had written: ""¦and Chu-Carroll corrected him."

Forgive my imprecision. For the purpose of showing Berlinski's calculation to be of no value (the correction that was offered), Chu-Carroll did not offer a correction of pre-biotic possibilities, because none was needed. He did include his own suspicion, but that was irrelevent to the point.

Bradford(2): Why aren't the two events completely independent?

I don't know if they are independent or are not, nor do Berlinski or Chu-Carroll. You have to prove independence before you multiply probabilities, or the calculation is meaningless.

Comment by One Brow — March 28, 2008 @ 12:12 pm

The Pixie Says:
March 28th, 2008 at 12:14 pm
Wrong. The only failing would be the math itself. The argument that OOL is bogus is on firm ground. Have you read the entire essay?

The essay has lots of arguments. I am specifically talking about the argument alluded to in the OP. That uses maths. If the maths fails, that specific argument fails.

Are we in agreement that that specific argument was wrong? Or are you only prepared to go on record as saying that you have not claimed to be impressed by the argument? What is it to be; are you going to stay on the fence, or actually take a position in this argument?

Brad: I understand that there is an increased likelihood of a second demon on Mars if we find the first one.

Pixie: Excellent. Now can see how why this is wrong?

Brad: Have you changed your mind? You indicated previously that "if we find one, the chances of finding another will be much higher."

No, I have not changed my mind. Why would you think that?

How are you doing with these (I know, hoping they will be fogotten):

p(two demons on Mars) = p(demon on Mars) x p(demon on Mars) p(template|self-replicator) = p(template)

Comment by The Pixie — March 28, 2008 @ 12:14 pm

Bradford Says:
March 28th, 2008 at 12:16 pm
OB: I don't know if they are independent or are not, nor do Berlinski or Chu-Carroll. You have to prove independence before you multiply probabilities, or the calculation is meaningless.

Fine. Is it your view that sufficient data exists on which we can base an assessment for the probability that a single self-replicating molecule would arise?

Comment by Bradford — March 28, 2008 @ 12:16 pm

Bradford Says:
March 28th, 2008 at 12:20 pm
Pixie: p(two demons on Mars) = p(demon on Mars) x p(demon on Mars) p(template|self-replicator) = p(template)

Pixie, we are well aware that you are quite the mathematician but do you have the slightest idea as to how a self-replicating RNA molecule would arise on prebiotic earth? There are no probability assessments in the absence of your ability to respond to that question.

Comment by Bradford — March 28, 2008 @ 12:20 pm

The Pixie Says:
March 28th, 2008 at 2:45 pm
Perhaps we could consider an alternative origins theory, YEC. YEC says mankind was created. I am not sure what the odds are of a creator, but let us say 50% for the sake of argument. So the probability that mankind was created is 50%. Now the important bit…

What about dogs? Well there is a 50% chance of a creator, so a 50% chance the dog kind was created. similarly for the cat kind, etc. I seem to remember a figure of 17,000 kinds altogether, each with a 50% probability. Overall then:

p = 0.5^17000 = 1 in 10^5100

Thus it is clear that the probability of the YEC scenario is 1 in 10^5100. Far, far less than what Berlinski calculates for abiogenesis.

Of course, they might be something wrong with my calculation (besides the estimate of 50%). Can anyone see what it is?

Comment by The Pixie — March 28, 2008 @ 2:45 pm

Rock Says:
March 29th, 2008 at 4:11 pm
A paper was cited in another topic that included this statistical argument (that I noticed is commonly repeated):

"Over 99% of all the species that have ever lived are now extinct"¦"

Except that >99% of all the species that have ever lived are known to live now, and >1% of all species are known to be extinct.

(For some unfathomable reason an "ateleological" argument.)

Never mind the a-/teleology; what about the statistics?

Comment by Rock — March 29, 2008 @ 4:11 pm

The Pixie Says:
March 29th, 2008 at 5:26 pm
Except that >99% of all the species that have ever lived are known to live now, and >1% of all species are known to be extinct.

Why do you believe that?

Comment by The Pixie — March 29, 2008 @ 5:26 pm

One Brow Says:
March 31st, 2008 at 11:56 am
Fine. Is it your view that sufficient data exists on which we can base an assessment for the probability that a single self-replicating molecule would arise?

I think we have enough information that, if we make a few dozen or hundred assumptions that are not verifiable, we can make a calculaiton based on those assumptions. Each person must decide for themself it that is an adequate assessment. Personally, I think the whole notion of applying probabilities post-hoc is somewhat flawed to begin with, but that's for another thread, I should think.

Comment by One Brow — March 31, 2008 @ 11:56 am

Get Real with Probability Assessments

59 Responses to “Get Real with Probability Assessments”

Featured Books