# Intelligent Design (pure)

#### Teax

##### huh?
milestone of the off-topic debate with WookieeB from http://intpforum.com/showthread.php?t=20961&page=4.

ID = Intelligent Design
but ID is often associated with creationism.
This thread is about - how reasonable/scientific is the version of ID that has nothing to do with creationism (pure ID so-to-speak), and why.

_________________________
argument for ID reconstructed from that other debate

CSI = Complex and Specified Information.
Complex = Unlikely to happen by chance as a random event.
Specified = necessary to function = if you take it out, the whole system becomes useless/stops working. irreducable complexity.
IC = Irreducably Complex

RM + NS = Random Mutation + Natural Selection (biological theory of evolution)
M + S = Mutation + Selection (mathematical theory behind evolution)

definition of ID:
ID is the search for evidence for ID_statement
ID_statement is the statement/assertion that only intelligence can produce CSI
ID_statement sais: if object has CSI then it has an intelligent designer

notice that this post does not disprove ID_statement, only disproves the popular search for evidence for ID_statement.

ID is made of 3 steps:
step_one:

try to make a case about how all known RM + NS theories cannot explain CSI (even remotely).

in other words: try to make a case that it's extremely unlikely or even impossible that evolution could ever produce CSI.

step_two aka "negative evidence":
if you believe ID is a valid hypothesis(=explanatory theory), then this argument sounds like this: "like with all competing hypothesis in science, if one hypothesis proves unlikely to explain a phenomenon, another hypothesis becomes more likely.".

another version of this argument:

an meta-induction (induction not over objects, but over theories/ideas of people) from step_one, to conclude that future theories will probably never explain CSI
ne_1: All known non-intelligent causes only explain objects without CSI
----this is the premise from step_one
ne_1c: An intelligent-designer-cause can explain objects with CSI
ne_2: For all non-intelligent causes, we infer(induce) from ne_1 that they probably don't explain objects with CSI
----no-bias assertion: <missing>... see below for explanation
ne_2b: All objects with CSI probably have an intelligent cause
----deduction from ne_2 and ne_1c

step_three aka "positive evidence":

pe_1: For all objects with CSI for which we found the cause, ID was always the cause
pe_1b: For all objects for which we found out that the cause was not ID, have no CSI
pe_2: For all objects with CSI, from pe_1 and ne_2b we infer(induce) that they probably have an intelligent cause.
----no-bias assertion: <missing>... see below for explanation

obviously the intelligent design of human life will automatically follow from pe_2

_________________________
flaws

an inference is a conclusion based on a premise

we often use the word 'infer'. but there are 2 basic kinds: deduction and induction.
to understand why ID is a bad argument, you have to first understand: when is induction not valid?

induction: (the no-bias assertion is explained here)
an induction is a form of probability argument based on sampling

premise = a sample = sampling space = representative subset of the population
conclusion = prediction about the whole population

induction only works if the premise is representative of the conclusion.
otherwise it's called a bias. if a sample is biased, inductive reasoning is not possible

example:
premise: we randomly pick 20 doors (=sample) in this house and count their locks. all these doors have 2 locks.
conclusion: we infer, all of the doors(=population) in this house have 2 locks.

let's assume the premise is true (we can count to 2, the doors really have 2 locks).
unlike a deduction, an induction is never just true. If it is reasonable to believe the induction, we usually say it's true colloquially, but it's implied that we actually have to call it "probable" or "likely". But it can always be disproven at any time with.....
• ....a counterexample. once the unknown becomes known, it can always disprove the induction.

example: once you find a door with 1 lock, the discussion is over.
.
• ....a bias. because an induction is an inference based on sampling, any bias that we discover in the sample challenges the induction.
For every challenge, the induction has the burden of proof to show that there is no bias. Otherwise the induction is not applicable to the whole population anymore, and is only applicable to the bias.

bias challenge: find a property of the population that you excluded in your sample. if you additionally find a correlation/dependence(=opposite of independence) between the sample and the excluded property, then the challenge cannot be refuted.
correction: if the challenge cannot be refuted, the induction can be rewritten to apply to only the bias, thus the bias disappears
defence: to refute the challenge, show that your sample is representative of the whole population. in other words: show that the bias is independent from the property that you make a prediction about. once you do that, this particular bias becomes irrelevant for the validity of the induction.

example:
bias challenge: I noticed you only counted the locks on non-red doors. (the random sample didn't include any red doors) this means you cannot infer that all doors have 2 locks, because we have no idea about the number of locks these red doors have.
correction: instead you can only infer that all non-red doors have 2 locks.
defence: every resident in this house is allowed to paint his door however he wants. the color of a door does not have any influence on the number of locks - because you can always repaint any door into any different color, without touching the number of locks, thus proving that the color of the door and the lock count are independent.

challenge has been refuted, the induction is still reasonable.

note that if a bias is known beforehand, making the induction would be pointless, because it would instantly be challenged. That is why it is necessary to assert that there is no (obvious) bias, everytime you make an induction. this is what I mean by "no-bias assertion" in the definition of the ID argument.

step_one: most interesting point. nothing wrong with questioning evolution theory.

step_two:
looking at ID_statement:
"only an intelligent cause can produce CSI"
this is not a hypothesis. It is a combination of 2 statements:
id_1. "intelligent cause can produce CSI"
id_2. "non-intelligent cause cannot produce CSI"

id_2 is not a hypothesis itself, it's a statement about other ones.
id_1 is a hypothesis, but it is not falsifiable and thus has no predictive power, and thus has no explanatory power. intelligent designer is applicable to everything, even not CSI objects, consequently - disproving other hypothesis, does not give this hypothesis any credence.

note - that doesn't mean that id_1 cannot be proven at all, it just means that it cannot be "proven"(=made reasonable to believe) by "disproving"(=made un-reasonable to believe) other hypothesis.

the other version of this argument is a meta-induction. It's an induction about theories and peoples' ideas, instead of real objects, which is not a valid area for induction. here's the induction in step_two:
• ne_1: All known non-intelligent causes only explain objects without CSI
• ne_2: For all non-intelligent causes -> infer(induce) from ne_1 that they don't explain objects with CSI
bias challenge:
• the sample is dependant(correlates) on time because the sample only contains theories discovered in the past.
• peoples' ideas are dependant on time
• peoples' ideas are dependant with the explanatory power of the theories.
• => hence the sample is dependant on the explanatory power of the theories. thus this challenge is irrefutable, proving that this induction is definetly wrong. We cannot infer anything about the future of theories from this.

this is the typical bias challenge for meta-induction attempts (inductions about peoples' ideas).
• correction: For all non-intelligent causes in the past, infer(induce) from ne_1 that they don't explain objects with CSI

note that this kind of meta-induction over theories has the same structure as the god-of-the-gaps-argument saying "since we cannot yet explain it, let's believe a hypothesis with zero explanatory power".

here's an example of a meta-induction:
A kid likes saying random numbers. he would just walk all day saying random numbers out loud.
premise: You start listening to the kid. the first 10 numbers are below 100000.
conclusion: you infer(induction) that all subsequent numbers will also be below 100000.

bad induction. the first 10 numbers are the earliest in time = they depend on time. the kid's ideas change over time = depend on time. thus, the numbers the kid will give in the future are dependent on time. We cannot infer anything in this situation through induction alone. This is the same default refutation for all meta-inductions (inductions about peoples' ideas).

step_three:
first, notice that pe_1b is equivalent to pe_1, so we can just skip it. step_two cannot be used here since it was not a valid argument. let me remove the refence to ne_2b, we check if step_three has any weight without step_two:

• pe_1: For all objects with CSI for which we found the cause, ID was always the cause
• pe_2: For all objects with CSI, from pe_1 we infer(induce) that they have an intelligent cause.
bias challenge: the sample only contains non-life. because all the object, for which we found the cause, were non-living objects.
correction: For all non-living objects with CSI, from pe_1a we infer(induce) that they probably have an intelligent cause.
notice that we use the corrected version of pe_2 everyday.

_________________________
conclusion: step_three adds nothing ontop of step two.
In other words: "positive evidence" is not an argument for ID, it is only as believable as the "negative evidence", so there's no point in even stating it.
In other words: the "negative evidence" support of ID remains unchanged, doesn't matter whether you add the "positive edivence" or not. but since "negative evidence" was faulty to begin with, we are left with step_one.

#### Teax

##### huh?
this is a direct reply to the latest points from http://intpforum.com/showthread.php?t=20961&page=4

WookieeB said:
I noticed a problem with the original argument. Unless I am wrong, (......)

the idea to split it into object1 + object2 is good. but the rest you said about the
parking analogy point to a faulty understanding of the analogy. I should have explained it better from the start. here are all the sub-analogies:
vehicle
= object with CSI

small vehicle
= constructed objects/non-life

big vehicle
= lifeform

4 wheels
= intelligently designed

non 4 wheels/more then 4 wheels
= not intelligently designed

this parking lot
= things we observe that they at least exist. we can not necessarily observe everything about them, like their wheelcount. but we can see they're there.

knowing how many wheels a vehicle has
= inspected object and found it's origin. for exaple things you made, or saw how they are made. you know for sure they're ID

not knowing how many wheels a vehicle has
= object we did not analyse. like Mt. Rushmore, you don't analyze it to be 100% sure who made it, you just infer.

here's a visual of a parking lot where you cannot see the wheels of some of the cars but you can see their sizes

=================================================
y1. all small vehicles in this parking lot, of which we know how many wheels they have, have 4 wheels.
y2. for all small vehicles in this parking lot, can we infer that they have 4 wheels? yes ok
y2b. for all big vehicles in this parking lot, can we infer that they have 4 wheels? ...... nah
=================================================

The fact that the small vehicles have 4 wheels tells us nothing about the big vehicles, except a suggestion that 4 wheels is a possibility. but the induction is biased, and inference for any degree of certainty not possible.

What does it mean that 4 wheels is a possibility: if you never saw the color blue, you would never suspect that blue objects even exist. If you saw green objects in the past, the color "green" becomes a possiblity for future objects.

if you saw big vehicles, but never saw their wheels, you maybe cannot imagine a big vehicle with 10 wheels, but that doesn't mean that the induction a y1 -> y2b becomes more valid.
note that the back wheels are double

WookieeB said:
The negative evidence didnt "prove" all vehicles have 4 wheels, because we still are allowing the possibility of there being a big vehicle with more than 4 wheels. It's just that such a vehicle has never been seen yet. The concept lives, but the demonstration of it does not...yet. Once someone builds the 10 wheeler and it is observed, the negative argument is not valid, then the inference fails.
ofc you didn't prove in the hardcore sence, it's an induction afterall, so the possibility for counterexamples is implied. Stating that your induction is refutable is redundant and thus doesn't make the induction any more valid.

also this quote makes it sound like the burden of proof is on the rest of the world to disprove the inference that big vehicles have more then 4 wheels. but it's not true, the inference was biased and thus faulty to begin with. The burden of proof is on you to show that your biased sampling is representative of the whole population of vehicles.

WookieeB said:
For 100% O AND bv, 0% O object 2 = NW
I agree, but this is also true
For 100% O AND bv, 0% O object 2 = 4W
because it actually should look like this
For 100% O AND bv, 0% O object 2

the only technologically advanced species you ever saw (humans) has hair. you cannot infer that all aliens also have hair, just like us, from the fact that you never saw an alien without hair. because you never saw any alien, hair or otherwise. (or did you? )

WookieeB said:
What is genetic code for the new function, then calculate probability of those mutations arriving by chance, factor in things like time, genetic drift, etc.
I gave you 2 options:
• retrospect_option: are you taking 2 species, A and B and calculating the probability for A to evolve into B (even if species A and B are only sliiightly different).
• openended_option: are you calculating the probability for A to evolve into any species that A could possibly evolve to, that has more CSI than A
you said that you took the openended_option, but you just described the retrospect_option. You are not allowed to just take any specific DNA of a specific function and assert that the probability for this specific function to evolve is ridiculously low, because I agree, it is, but it doesn't prove/disprove anything.

Instead you have to compute the probability for any DNA that would produce any new CSI. Otherwise you have unjustified assumptions (unless you show why it's not) that this particular functionality or this partilucar DNA was the only CSI that could have evolved at that point.

I'll rewrite the "dice result 3" analogy that you completely misunderstood:
we have a random number generator between 0 and 10^100000.
retrospect_option: once you draw a random number, let's say it's 15313425691932, you discover that wow, it's dividable by 81. the probability of the outcome 15313425691932 is unimaginably small. but that doesn't prove anything.
openended_option: before drawing a specific random number, the probability of a random number to be dividable by 81, is about 1%. no big deal.

WookieeB said:
just checking, do you believe the eye could have evolved, but unlikely? or are you saying that evolution cannot account for any eye in any creature whatsoever and hence it's impossible?

WookieeB said:
This is true. It may depend on the process under consideration, for some of them it is impossible (*ie: a purely law based process cannot cover any complexity requirement, chance cannot cover the specificity) *=late edit and correction
could you give an example for what is possible and what is impossible?

also what do you mean by
"purely law based process cannot cover any complexity requirement"
and why do you believe this?

#### Madoness

##### that shadow behind lost
Step 2 is flawed with a monty hall problem. Step 1s conclusion could be only valid if we were all to close our eyes and ears and never accept new information.

These steps taken altogether are not valid because it its not taken as a possibility of unknown proportions.

Taken purely scientifically. we should go with the evidence. These "steps" here have little or no value at all.

#### Cognisant

##### Prolific Member
In the worlds oldest religions (shamanism, paganism, etc) creation stories were largely devoid of intelligent design, instead people believed everything came into being as the result of some mystical dream era or the world is the corpse of some mighty slain being or life was the result of the earth and sky spirits getting their groove on.

It was only later when people started to cultivate farms, pave roads and build towns that the creationism became a theory people could accept as plausible.

#### higs

##### Omg wow imo
Complexity could arise from chaos and infinite possibilities, stability acquired as the system grows and retro acts on itself.

#### Brontosaurie

##### Banned
step everything: stifle reason with rigid technicality, win "debate" as nobody bothers to challenge vague cryptical assertions

why this obscure style of argument if you're against ID?

#### Cherry Cola

##### Banned
I think the only one who might be able to reply to this is WookieB, no one else can possibly understand it. Maybe it's not that hard but when you look at it it's like naaaah, fuck this shit :P

#### Teax

##### huh?
step everything: stifle reason with rigid technicality, win "debate" as nobody bothers to challenge vague cryptical assertions

why this obscure style of argument if you're against ID?

I think the only one who might be able to reply to this is WookieB, no one else can possibly understand it. Maybe it's not that hard but when you look at it it's like naaaah, fuck this shit :P

haha yes very good point, I asked myself exactly that question. It seems like something that was initially so obvious because intuitive, turned out to be unconvincing to another INTP. so I kept diggin to find the root cause of why it is he still believed in ID despite being so logically grounded.

This eventually lead to that hunk of ridiculously structured text in post #1, which is actually "self explanatory" and could be stated in like 3 or 4 lines of text, but for someone who's not convinced, I tried to cover all bases.

In retrospect I was just procrastinating something more important....

#### WookieeB

##### Active Member
argument for IDreconstructed from that other debate

Good idea to move it to another thread. I was feeling guilty for somewhat derailing the other one.

In general, I congratulate you for the detail and volume of the breakdown of the argument. Unfortunately, I disagree with how you presented it, and it basically all comes down to one word that you use....

ID_statement is the statement/assertion that only intelligence can produce CSI

You even italicized it yourself. "Only" doesnt belong there.

If you insist on that word, then you have to add another adjective to clarify the argument.

ID is the assertion 1) that for objects with CSI that are of known origins, that origin has "only" been intelligence. and 2) for known non-intelligent causes, they cannot produce objects with CSI within a reasonable probability (thus not necessarily impossible, but extremely unlikely). So with these premises, for objects with CSI of unknown origin, the best explanation (inference) of that origin is an intelligent agent.

It thus leaves open a currently unknown non-intelligent cause for CSI objects. But assuming such a cause, of course once that cause becomes known, it would break the premise and the inference would fail.

ID is made of 3 steps:
step_one:...

No problem here, though this step is probably not needed. It fits under step 2 as a specific example.

The subject of this step is where the greatest debate over ID usually happens, not so much with the structure of IDs argument itself.

step_two aka "negative evidence":......

Mostly OK, though I would change ne_1 somewhat. An intelligent cause can produce objects without CSI OR with CSI, but known non-intelligent causes cannot produce objects with CSI. That "only" word is getting in the way again. So -

ne_1: All known non-intelligent causes cannot explain objects with CSI

step_three aka "positive evidence":

i have no problem here with how you stated it.

As for the "Flaws" section, it basically boils down to the same problem I mentioned above.

From step_two -
"only an intelligent cause can produce CSI"
this is not a hypothesis. It is a combination of 2 statements:
id_1. "intelligent cause can produce CSI"
id_2. "non-intelligent cause cannot produce CSI"

Take out the word "only" in the first phrase. For the broken down id_1 and id_2, insert the word "known" at the beginning.

This, as the negative argument, allows for an undiscovered non-intelligent process to come and save the day, or allows one of the known non-intelligent processes to step up an refute id_2. Either way, this makes the argument falsifiable.

But until that happens, the inference stands.

conclusion: step_three adds nothing ontop of step two.
In other words: "positive evidence" is not an argument for ID, it is only as believable as the "negative evidence", so there's no point in even stating it.
In other words: the "negative evidence" support of ID remains unchanged, doesn't matter whether you add the "positive edivence" or not. but since "negative evidence" was faulty to begin with, we are left with step_one.

Since we have clarified the argument, especially the negative portion, the above conclusion no longer holds true.

#### WookieeB

##### Active Member
the idea to split it into object1 + object2 is good. but the rest you said about the
parking analogy point to a faulty understanding of the analogy. I should have explained it better from the start. here are all the sub-analogies:
*spoiler*
vehicle
= object with CSI

small vehicle
= constructed objects/non-life

big vehicle
= lifeform

4 wheels
= intelligently designed

non 4 wheels/more then 4 wheels
= not intelligently designed

Ok, thanks for the clarification, but with this I do not think it makes a very good analogy for ID. Essentially ID is saying that there are currently no known non-intelligent causes that could produce a lifeform. Following your analogy-objects that means we do not know (and cannot know by direct observation) how many wheels a big vehicle has. It also means that we have never yet observed any vehicle (big or small) with more than 4 wheels.

A vehicle (big or small) with more than 4 wheels is no more than an idea, it is not instantiated in any physical reality, and the inference is that is is highly unlikely to ever be reality, if even possible.

Sorry, but it's a bad analogy.

also this quote makes it sound like the burden of proof is on the rest of the world to disprove the inference that big vehicles have more then 4 wheels. but it's not true, the inference was biased and thus faulty to begin with. The burden of proof is on you to show that your biased sampling is representative of the whole population of vehicles.

No. See my response above. The problem with the analogy is we are saying that there are big vehicles, but we cannot see how many wheels they have, like they are completely covered in panels. Plus, we have never seen a vehicle with more than 4 wheels, ever, so any talk of a vehicle with more than 4 wheels is currently conjecture.

It doesn't relate well to the ID argument.

I agree, but this is also true
For 100% O AND bv, 0% O object 2 = 4W
because it actually should look like this
For 100% O AND bv, 0% O object 2

the only technologically advanced species you ever saw (humans) has hair. you cannot infer that all aliens also have hair, just like us, from the fact that you never saw an alien without hair. because you never saw any alien, hair or otherwise. (or did you? )

We have no evidence of aliens. Thus to argue for them is really a 'cause-of-the-gaps' argument.

I gave you 2 options:
• retrospect_option: are you taking 2 species, A and B and calculating the probability for A to evolve into B (even if species A and B are only sliiightly different).
• openended_option: are you calculating the probability for A to evolve into any species that A could possibly evolve to, that has more CSI than A
you said that you took the openended_option, but you just described the retrospect_option. You are not allowed to just take any specific DNA of a specific function and assert that the probability for this specific function to evolve is ridiculously low, because I agree, it is, but it doesn't prove/disprove anything.

No, I get it. I'm willing to take the open-ended option, but you have to realize it cannot stay open-ended forever, otherwise what are we trying to explain. At some point, you have to show what species A 'evolves' into, whether that is species B, C, D....ZZZZZZ. You cannot just say species A changes and leave it at that. ID doesnt say that species A cannot ever change. It concedes that a process like RM+NS exists. What is at question is how far can RM+NS take something? ID says that is very limited, that there is an edge to how far species A can change, and anything beyond that edge would most likely fall to an intelligent cause. You cannot make any judgement on any of this until you define what has changed.

I'll rewrite the "dice result 3" analogy that you completely misunderstood:
we have a random number generator between 0 and 10^100000.
retrospect_option: once you draw a random number, let's say it's 15313425691932, you discover that wow, it's dividable by 81. the probability of the outcome 15313425691932 is unimaginably small. but that doesn't prove anything.
openended_option: before drawing a specific random number, the probability of a random number to be dividable by 81, is about 1%. no big deal.

I agree, no big deal. But you have not defined any specificity or meaning to the number that is chosen. Any number from 0 to 10^100000 is possible, and each of the possibilities is as probable and any other one. But big deal. This is all Shannon information theory.

What is important is the specificity. If you want to specify that the number has to be wholly divisible by 81, that is good. You defined a probability target/meaning. But that is a pretty high probability, a little better than 1 in 10^2. But now lets make a probability of 1 in 10^150. You will still have quite a few hits within the range you defined. So now we have to correspond your range to something real, like time. 1 in 10^150 relates to every single atom in the known universe, the maximum rate per second at which transitions in physical states can occur (10^45), and billion times longer than the age of the universe. That is the probability ID proper measures against; if you cannot beat that probability, then it is not likely to ever happen.

just checking, do you believe the eye could have evolved, but unlikely? or are you saying that evolution cannot account for any eye in any creature whatsoever and hence it's impossible?

Well, short of saying that anything is impossible, I am saying that the eye evolving is "unlikely". But soooooooooo unlikely that it is effectively impossible.

could you give an example for what is possible and what is impossible?

also what do you mean by
"purely law based process cannot cover any complexity requirement"
and why do you believe this?

I realized later that I sort of screwed up even my edit. So I'd take back most of that statement if I could. Something law-like would be a property of physics, like crystal formation. It is usually very specified, but the complexity is low (ie a high enough probability of occurring that I wouldnt argue against it could happen). The problem with law-like process though is that the specificity is locked in, it never changes. Crystal formations always follow the same process/rules. That is not saying that crystals may appear different at a macro level (like different snowflakes which are composed of water crystals), but the underlying structure is always the same (how the molecules connect based upon the environmental conditions)

#### Jennywocky

##### guud languager
Well, short of saying that anything is impossible, I am saying that the eye evolving is "unlikely". But soooooooooo unlikely that it is effectively impossible.

Could you explain more on why you think this?

#### redbaron

##### irony based lifeform
Well, short of saying that anything is impossible, I am saying that the eye evolving is "unlikely". But soooooooooo unlikely that it is effectively impossible.

People still use the, "it's too complex to have occurred naturally!" argument?

#### Cognisant

##### Prolific Member
Well, short of saying that anything is impossible, I am saying that the eye evolving is "unlikely". But soooooooooo unlikely that it is effectively impossible.
We have a blind spot on our retina, an obvious and completely unnecessary design flaw which we can't get rid of because by chance it evolved that way and we can't unevolved to re-evolve a different way.

Maybe the planet was seeded with genetically engineered bacteria billions of years ago so in that sense maybe we are the product of intelligent design, but the human body certianly wasn't designed by any intelligent being and fossil records indicate all other complex organisms on our planet evolved from the same primordial origins.

#### WookieeB

##### Active Member
Could you explain more on why you think this?

Evolution (RM + NS) can work fine when one small step along an the way gives an advantage, which would usually correspond to a single and specific point mutation in DNA. ID has no problem with this.

But the problem comes when there many steps, or many mutations, or many parts, necessary to gain some advantage? And I would hold the same view of many biologists, that the eye would fall into a category that requires multiple parts, each with their own function, arranged and coordinated together to perform a higher function (ie, vision).

The usual narrative that I hear is similar to a story Richard Dawkins (and even Darwin himself) tells about starting with a light-sensitive spot, then it finds itself in a depression to allow some sense of direction, then the depression concaves more, which progresses to an almost enclosed space with a pin-hole opening for light, to a transparent covering, to a lens, etc....

That anatomical story sounds all nice and simply progressive, but it really is missing TONS of detail. Go back to the beginning....."Start with a light-sensitive spot"... and immediately put on the brakes. That in itself is a HUGE assumption. What is involved with even this simple staring point? To quote from Michael Behe -
Here is a brief overview of the biochemistry of vision. When light first strikes the retina, a photon interacts with a molecule called 11-cis-retinal, which rearranges within picoseconds to trans-retinal. The change in the shape of retinal forces a change in the shape of the protein, rhodopsin, to which the retinal is tightly bound. The protein's metamorphosis alters its behavior, making it stick to another protein called transducin. Before bumping into activated rhodopsin, transducin had tightly bound a small molecule called GDP. But when transducin interacts with activated rhodopsin, the GDP falls off and a molecule called GTP binds to transducin. (GTP is closely related to, but critically different from, GDP.)

GTP-transducin-activated rhodopsin now binds to a protein called phosphodiesterase, located in the inner membrane of the cell. When attached to activated rhodopsin and its entourage, the phosphodiesterase acquires the ability to chemically cut a molecule called cGMP (a chemical relative of both GDP and GTP). Initially there are a lot of cGMP molecules in the cell, but the phosphodiesterase lowers its concentration, like a pulled plug lowers the water level in a bathtub.

Another membrane protein that binds cGMP is called an ion channel. It acts as a gateway that regulates the number of sodium ions in the cell. Normally the ion channel allows sodium ions to flow into the cell, while a separate protein actively pumps them out again. The dual action of the ion channel and pump keeps the level of sodium ions in the cell within a narrow range. When the amount of cGMP is reduced because of cleavage by the phosphodiesterase, the ion channel closes, causing the cellular concentration of positively charged sodium ions to be reduced. This causes an imbalance of charge across the cell membrane which, finally, causes a current to be transmitted down the optic nerve to the brain.

Within that description are quite a few protiens and molecule "parts" that you have to explain the existence of (within this environment), timing of appearance, and coordination with other parts. And for each additional part you have, the probability goes up exponentially (its multiplicative, not additive). This scenario also does not even consider what has to occur within brain (the electrical signal has to be translated to an understandable format) which is it's own sense is a gargantuan feat to develop by random processes (and I'm even here conceding the existence of a brain).

And even if you want to try to say that there might be multiple ways of accomplishing the function of the quoted text, you are still left with explaining probabilities of random processes developing them. Plus there are limits as to how far you can realistically go beyond the process described above.

And this is just for determining a starting point for the eye.

As you throw more and more changes in, you have to account for the large informational changes at the molecular level that would be necessary, each step likely translating to immensely high probabilities.

And then even beyond that, you might not even have the necessary tools to accomplish the job. Genetic changes at the DNA/AA level may account for the appearance of the base parts (proteins), but recent discoveries seem to indicate that the spatial arrangement of those parts is not controlled by DNA, but rather by epigenetic factors (which themselves are just as specified and complex, if not moreso, as DNA).

#### WookieeB

##### Active Member
We have a blind spot on our retina, an obvious and completely unnecessary design flaw which we can't get rid of because by chance it evolved that way and we can't unevolved to re-evolve a different way.

Saying that it is a design flaw would still be conceding design, ie an intelligent cause. And whether or not it really is a design "flaw" is a matter of subjective tastes. It appears that it actually is a necessary part of vertebrate eyes. Competing requirements and trade-offs are actually a hallmark of design and wouldnt be something you would NOT expect to see in a darwinian process. Besides, the eye easily compensates for it, using a combination of movement (nystagmus) and the brain's 'software' compensating.

but the human body certianly wasn't designed by any intelligent being and fossil records indicate all other complex organisms on our planet evolved from the same primordial origins.

Subjective assertion. fortunately the data doesnt support it.

#### Teax

##### huh?
and it basically all comes down to one word that you use....You even italicized it yourself. "Only" doesnt belong there.
without it the flaws stay the same:

step_three is irrelevant, regardless of whether the other steps are true or false. It can be just left out without hurting ID argument.

the only thing the 'only' does is add id_2. without 'only', id_2 disappears, and it makes ID easier to negate, because it makes ID_statement obviously-unfalsifiable. That's why I included 'only', remember you said:
ID is falsifiable. (...). If a non-intelligent cause can be demonstrated able to produce CSI, then ID would fail.
if that demonstration would falsify ID_statement, it means ID_statement's claim is that objects with CSI must have an intelligent cause, which is equivalent to "only intelligence can produce CSI"

If you insist on that word, then you have to add another adjective to clarify the argument.

ID is the assertion 1) that for objects with CSI that are of known origins, that origin has "only" been intelligence. and 2) for known non-intelligent causes, they cannot produce objects with CSI within a reasonable probability (thus not necessarily impossible, but extremely unlikely).
(...)
Take out the word "only" in the first phrase. For the broken down id_1 and id_2, insert the word "known" at the beginning.
this is not the statement that ID tries to argue for, it is the argument. I already included all those points by saying
"ID is the search for evidence for ID_statement"
the statement is merely the thing you are trying to prove that it is reasonable to believe

It thus leaves open a currently unknown non-intelligent cause for CSI objects. But assuming such a cause, of course once that cause becomes known, it would break the premise and the inference would fail.
ofcourse, no induction can be 100% proven. (as I explained in OP). this quote does not add credibility to ID_statement however - because it does not show that ID_statement is falsifiable - it only shows that this particular argument that tries to support ID_statement is falsifiable.

example:
assume we have a vase.
broken vase theory sais: (analogy to ID_statement)
"the vase fell and scattered"
stoned kid evidence sais: (analogy to ID)
"that vase was broken because a stoned kid pushed it"
if you found evidence that the kid wasn't stoned, or wasn't even an actual kid, you negate the evidence not the theory. the theory is a separate thing! please don't mix these 2.

the broken vase theory is falsifiable by e.g. finding out that it's not broken. so it's not a very good analogy for ID_statement, but it illustractes this point nicely

No problem here, though this step is probably not needed. It fits under step 2 as a specific example.

The subject of this step is where the greatest debate over ID usually happens, not so much with the structure of IDs argument itself.

yeah I just wanted to show you that even if we assume step 1, just for the sake of argument and fun and Ti training, the other 2 steps still would not be valid.

Mostly OK, though I would change ne_1 somewhat. An intelligent cause can produce objects without CSI OR with CSI, but known non-intelligent causes cannot produce objects with CSI. That "only" word is getting in the way again. So -

ne_1: All known non-intelligent causes cannot explain objects with CSI
I agree, it's equivalent afterall, but your version is more to the point.

Ok, thanks for the clarification, but with this I do not think it makes a very good analogy for ID. Essentially ID is saying that there are currently no known non-intelligent causes that could produce a lifeform. Following your analogy-objects that means we do not know (and cannot know by direct observation) how many wheels a big vehicle has. It also means that we have never yet observed any vehicle (big or small) with more than 4 wheels.
yes! now were talking about the same thing here
note that we have observed objects with 6 wheels, just not on vehicles
6 wheels.. sorta... let's call it a 'toy'?

A vehicle (big or small) with more than 4 wheels is no more than an idea, it is not instantiated in any physical reality, and the inference is that is is highly unlikely to ever be reality, if even possible.

Sorry, but it's a bad analogy.
(...)
No. See my response above. The problem with the analogy is we are saying that there are big vehicles, but we cannot see how many wheels they have, like they are completely covered in panels. Plus, we have never seen a vehicle with more than 4 wheels, ever, so any talk of a vehicle with more than 4 wheels is currently conjecture.
anything said about the wheels of big vehicles is conjecture, not just the existence of non 4 wheels. the way I see it you just demonstrated why it's a great analogy - because you missed the point about that the induction is not possible at all.

we know that objects need to have wheels by necessity(=all objects have to have a cause), and we have seen that toy with 6 wheels(=we know objects with natural cause), so we can imagine big vehicles to have not 4 wheels aswell(=we can imagine objects with CSI to have natural cause). but without theory of evolution(=assuming youre right and evolution cannot account for CSI) it's all conjecture we cannot induce anything. and certainly not ID.

you cannot induce that the big vehicles have 4 wheels because you only have small vehicles to begin with, your sample is biased (see OP post for explanation of induction bias).

No, I get it. I'm willing to take the open-ended option, but you have to realize it cannot stay open-ended forever, otherwise what are we trying to explain. At some point, you have to show what species A 'evolves' into, whether that is species B, C, D....ZZZZZZ. You cannot just say species A changes and leave it at that.
Go back to the beginning....."Start with a light-sensitive spot"... and immediately put on the brakes. That in itself is a HUGE assumption. What is involved with even this simple staring point?
you're still missing the open-ended-option. you didn't consider the many other potential traits a species might have gotten to (later)evolve into something completely different. you assumed this specie's DNA is the trait that it developed to produce this machine. but why do you assume this machine in this very form is the machine that was going to be produced in the first place? At the moment you are only allowed to claim that this specific CSI is unprobable, and we all agree on that, just like the random number generator example.

for those who didn't see it:
http://www.youtube.com/watch?v=Nwew5gHoh3E
most relevant part starts at 4:55, where he hints at the different paths evolution can take.

That is why dawking's explaination is so valid, but missing one point that only more advanced people will understand. the 'light sensitive spot' did not have to appear at any particular time, or at any particular position, or with any particular mechanism/chemicals that is able to register light. It simply has to appear, maybe even several times (what use is a light sensitive spot on the inside of the primitive-fish for example). once it appears in a place that it proves useful in, that trait will establish itself like meme, converting all the species with this new trait, simply by NS. fish have the advantage of producing thousands of offspring making the propagation rate for any useful trait extremely high relative to big land animals.

example: depending on new CSI that comes along, monkey predecessor-eye might have evolved into a stalk-eye/snail eye, but it didn't. the fact that it didn't cannot reduce the probability of having evolved CSI reduced in your calculation. because you wanted to show that evolution is unlikely to produce any new CSI, not just the particular one you're talking about. You are not allowed to exclude that monkeys could have stalk-eyes from your calculation, or any other potential course of evolution from the probability calculation, because only then you are allowed to claim that it would be unprobable to gain any CSI.

The problem with law-like process though is that the specificity is locked in, it never changes.
But the problem comes when there many steps, or many mutations, or many parts, necessary to gain some advantage? And I would hold the same view of many biologists, that the eye would fall into a category that requires multiple parts, each with their own function, arranged and coordinated together to perform a higher function (ie, vision).
in crystals ok. in evolution, you have NS - your perfect specificity selector, that chooses the right specificity at the right time and keeps evolving new useful traits. just not your specificity.

let's keep those unimaginative biologists out of the play here . Are those multiple parts needed to perform the exact specific same function it does now? yes. any function? no.

NS does not care about what you call a "higher function". as long as something is beneficial, it will be selected. the eye probably had the same main function all the time, not because the parts that the eye is made-of somehow support sight, but because sight is a far more useful feature of that thing, placed in all our predecessors, than any other possible use for the eye. But it's still possible that the very very first eye evolved from an ear or some antennae on the head of the premordial fish.

http://www.youtube.com/watch?v=a_5FToP_mMY
=== relevant part starts at 5m33s ===

the bacterial flagellum is a locomotion machine. it however has the same structure like a poison injector on other bacteria, and maybe evolved like a poison injector for a while untill a mutation made it movable at all, even slightly! and then NS took over and realized that locomotion was more beneficial use for this thing than poison injection.

basically the assumption of irreducable complexity is simply false. when you remove parts of the machine, the machine stops being useful for that one thing, that is true, but it's still useful for another.

the problem that evolution opponents have is that they assume one singleminded unalterable purpose for the trait. which is basically a circular argument they unknowingly commit into when arguing for ID, by assuming one single purpose and then deducing intelligence that is required to have that purpose .

#### WookieeB

##### Active Member
without it the flaws stay the same:

step_three is irrelevant, regardless of whether the other steps are true or false. It can be just left out without hurting ID argument.

Did you perhaps mean step_one can be left out? That I could understand. But step_3 is crucial, you cannot leave it out.

the only thing the 'only' does is add id_2. without 'only', id_2 disappears, and it makes ID easier to negate, because it makes ID_statement obviously-unfalsifiable.
This does not make any sense. If you take away id_2 you are left only with "id_1. "intelligent cause can produce CSI"", which by itself is not making any case. The "only" in your prefaced statement is wrong, so just chuck the ""only an intelligent cause can produce CSI" statement altogether. Use the 2+1 following statements in combination:
id_1. "known intelligent causes can produce CSI"
id_2. "known non-intelligent causes cannot produce CSI"
id_3. "in case of CSI object of unknown origin, the best explanation for origin is intelligent cause.

Now the step_two and step_three argument is falsifiable if you encounter an object with CSI of unknown origin. It is not making an "only" or "is" statement with regards to the origin being ID, but is saying that among available origins (ID or non-ID), ID is the best explanation.

That's why I included 'only', remember you said:
"ID is falsifiable. (...). If a non-intelligent cause can be demonstrated able to produce CSI, then ID would fail. "
if that demonstration would falsify ID_statement, it means ID_statement's claim is that objects with CSI must have an intelligent cause, which is equivalent to "only intelligence can produce CSI"
No, that is not what ID is saying. There is no "must" at all. It is saying that among possible explanations (ID and non-ID, including whatever sub-categories those may entail), according to our uniform experience the best explanation of the possibilities is ID.

this is not the statement that ID tries to argue for, it is the argument. I already included all those points by saying
"ID is the search for evidence for ID_statement"
the statement is merely the thing you are trying to prove that it is reasonable to believe
No.

The statement(s) ID is making are:
id_1. "known intelligent causes can produce CSI"
id_2. "known non-intelligent causes cannot produce CSI"
These are both supported by empirical evidence

The argument ID is making is:
id_3. "in case of CSI object of unknown origin, the best explanation for origin is intelligent cause.

It is NOT: "ID is the search for evidence for ID_statement", or to expand that statement: "Intelligent Design is the search for evidence of [only an intelligent cause can produce CSI]

ID is not making an "only" statement in any sense. The "only" statement may in fact be true, but ID is not saying that. It purposefully is acknowledging the unknown and that future discoveries may force a change/negation of id_1 and id_2. But until then, the inference to the best explanation stands.

ofcourse, no induction can be 100% proven. (as I explained in OP). this quote does not add credibility to ID_statement however - because it does not show that ID_statement is falsifiable - it only shows that this particular argument that tries to support ID_statement is falsifiable.
No, because your ID_statement is not what ID is saying. ID is falsifiable because it allows the possibility of something currently unknown to come along and prove it wrong.

#### Teax

##### huh?
Did you perhaps mean step_one can be left out? That I could understand. But step_3 is crucial, you cannot leave it out.
no I meant exactly that.

the "flaws" section, step_three in the OP post. it explains why step_three is an argument which by itself, is not a valid induction, and only as believable as step_two, if step_two were believable at all.

This does not make any sense. If you take away id_2 you are left only with "id_1. "intelligent cause can produce CSI"", which by itself is not making any case.
bingo. that is exactly what I'm saying, that the conclusion of ID (=ID_statement) is a poinless thing. ID is trying to find evidence for: "only intelligent cause can produce CSI"

The "only" in your prefaced statement is wrong, so just chuck the ""only an intelligent cause can produce CSI" statement altogether. Use the 2+1 following statements in combination:
id_1. "known intelligent causes can produce CSI"
id_2. "known non-intelligent causes cannot produce CSI"
id_3. "in case of CSI object of unknown origin, the best explanation for origin is intelligent cause.
No, because your ID_statement is not what ID is saying. ID is falsifiable because it allows the possibility of something currently unknown to come along and prove it wrong.
you misunderstood what ID_statement is. it's the final statement, the conclusion, the thing ID is trying to prove in the very end. forget about all the steps that lead up to it.
you're trying to again mix the argument/evidence with the conclusion.

"known intelligent causes can produce CSI" <- thats a premise not a conclusion.
"known non-intelligent causes cannot produce CSI" <- again, premise.
"in case of CSI object of unknown origin, the best explanation for origin is intelligent cause." <- this is the only conclusion. and that's exacly what id_1+id_2 is saying.

reminder:
id_1. "intelligent cause can produce CSI"
id_2. "non-intelligent cause cannot produce CSI"
id_1 + id_2 = ID_statement = "if object has CSI then it has an intelligent cause"

do you understand that:
the statement a theory makes, is the conclusion of some evidence/reasoning, otherwise the statement is not supported by reason/evidence.

ID_statement must be the conclusion of ID, otherwise theres no evidence/reasoning to support it.

Now the step_two and step_three argument is falsifiable if you encounter an object with CSI of unknown origin.
true but falsifiability of evidence does not falsify the conclusion.

It is not making an "only" or "is" statement with regards to the origin being ID, but is saying that among available origins (ID or non-ID), ID is the best explanation.
that is the same thing.

relativity theory is the best explanation for some of the phenomena. untill something better comes up. but you don't have to explicitely say that, it's implied.

relativity theory does not sound like this:
"based on the evidence that supports this, this is the best explanation for these phenomena, and that's why things behave like blabla and gravity blablah"
instead is sounds like
"blablah and gravity blahblah"
the rest is implied. search for evidence for relativity theory is independent from the statement that the theory makes.

ID is not like other theories/hypothesis. ID is not a theory. it's the evidence/reasoning for a theory. ID and natural selection are not "2 things of the same type", they're like "apples and oranges"

#### redbaron

##### irony based lifeform
WookieB said:
But the problem comes when there many steps, or many mutations, or many parts, necessary.

You're a fucking idiot.

#### Lot

##### Don't forget to bring a towel
You're a fucking idiot.

Very strong argument
So complex
Thought provoking
Wow
Mind blown
Such dead god
wow

#### redbaron

##### irony based lifeform
WookieB's argument has been repeated ad infinitum by idiot creationist/ID proponents. It's just dead wrong. The only thing required for, "more" complexity in evolution is time.

There is literally no difference whatsoever between a simple or an advanced mutation because the process occurs over time. Evolution has been observed first hand, in almost (if not more by now since it was years ago that I sifted through them) one hundred separate instances utilizing various animals. Reptiles, mammals, birds, insects and bacteria. Some of these evolutions involved growing entirely new body parts, in only a matter of a few years in short-lived species who progress through each generation quickly.

The only difference between the evolution of an extra body part and a complex structure like the eye? Time.

Since life on Earth is 3.5 billion years old, there's been ample time. Especially since generational periods of early animals are even less than humans. Here's some basic math for you:

In 30-40 generations species have been observed to grow entirely new limbs.

Humans have existed for 200,000 years, with a generational period of typically 20 to 23 years. That's 8,000 generations if we assume 25. That's only in one mammal with a long generation period. What about bacteria which existed 3.5billion years ago, with a generation time of hours?

In on year there's ~8800 hours. At a conservative guess that means 17600 generations of bacteria per year. 17600 generations, multiplied by 3.5 billion. Feel free to calculate that and just try to comprehend the sheer size of the number.

They eye isn't, "too complex" for evolution.

~

However it's not going to matter who or how many times this is demonstrated because idiots don't want to hear the truth, so they're just going to cling to their idiot ideas rooted entirely in either their inability or unwillingness to understand basic scientific concepts.

Disgustingly ignorant.

#### WookieeB

##### Active Member
anything said about the wheels of big vehicles is conjecture, not just the existence of non 4 wheels. the way I see it you just demonstrated why it's a great analogy - because you missed the point about that the induction is not possible at all.

I think you are missing a point in your analogy. In your analogy, every object in existence has wheels. (that is why in part I do not think it is a good analogy, it forces a person to think in a non-realistic manner to describe a realistic situation that is still under consideration. An analogy is supposed to simplify things or at least relate a subject to a more understandable experience). Yes, the wheel count of big vehicles is unknown. But the wheel count of small vehicles is known, and to our experience they ALL have 4 wheels. We now are presented with something (big vehicles) that bear many similar properties to small vehicles to which we know the wheel count. We know it has wheels, but we cannot see the wheel count on the big vehicles, but as we can relate much of it to small vehicles, and we have never seen a small vehicle with not 4 wheels, it is not a stretch to infer that big vehicles have 4 wheels too. That is not a "4 wheels for sure" statement, but rather a "If it looks like a duck, smells like a duck, quacks like a duck, then its probably a duck" statement.

we know that objects need to have wheel ... but without theory of evolution(=assuming youre right and evolution cannot account for CSI) it's all conjecture we cannot induce anything. and certainly not ID.
you cannot induce that the big vehicles have 4 wheels because you only have small vehicles to begin with, your sample is biased (see OP post for explanation of induction bias).
Up to the bold text you were doing fine. Without a theory of evolution, you can definitely induce something.
If in our experience, all vehicles where we can see the wheels are small vehicles, and the wheel count has always been 4, and we have never yet seen or heard of a real vehicle without 4 confirmed wheels. And now we see something that resembles a small vehicle, but that is just bigger, but it has it's wheels hidden. So, despite all our experience with wheels + vehicles, we cannot make an educated guess? We cannot link to our experience of small vehicles and only 4 wheels and infer that a similar object (but bigger) does not also have 4 wheels? Especially when we have been given no realistic reason to think that it may have other than 4?

you're still missing the open-ended-option. you didn't consider the many other potential traits a species might have gotten to (later)evolve into something completely different.
It's nice to muse about what something potentially might have alternately developed into, but that is all just wishful thinking, made-up fantasy, there is no evidence of it existing. I'm dealing with what IS existing and trying to explain that using known processes now in existence. Evolution doesnt cut it to explain how what IS came to be. That is the ultimate point of the negative ID argument.

And even if you still insist on considering the open-ended object, you are assuming the very process that you are trying to explain, that is evolution. I am asserting that evolution cannot make what we see exists, and you are not defending that evolution can make it, but are countering that evolution maybe can make something else that we have no evidence for. Your argument is a evolution-of-the-gaps proposal.

you assumed this specie's DNA is the trait that it developed to produce this machine. but why do you assume this machine in this very form is the machine that was going to be produced in the first place? At the moment you are only allowed to claim that this specific CSI is unprobable, and we all agree on that, just like the random number generator example.
No assumptions are being made, but I am dealing with what we have, what is real. If we know the starting point, we can determine probabilities on what resulted, assuming evolution. And because the vast majority of functional changes in proteins require multi-point mutations on DNA, and based on empirical studies that can determine the rate of viable possibilities of multi-point changes, assuming evolution, the probability gets very low. In other words, we can make an educated guess on the likelyhood of your "open-ended-options"

That is why dawkin's explanation is so valid, but missing one point that only more advanced people will understand. the 'light sensitive spot' did not have to appear at any particular time, or at any particular position, or with any particular mechanism/chemicals that is able to register light. It simply has to appear,
...but how did it appear? That is the question?!? You and Dawkins are doing the same thing and assuming what you are trying to prove. You are begging the question. It doesn't have to appear! You are only saying that because it exists. But evolution has no forethought, no plan, it doesnt have to do anything. You cannot assume the process just because the object you are trying to explain the cause of exists.

Dawkins description of Mt Improbable is assuming a fitness landscape that is always uphill up toward whatever evolutionary property he is talking about (nautilus eye), and assuming that each peak jump is a minor probability event (like a single mutation). But that is not a realistic scenario for most functional changes. There would be valleys in between different fitness peaks, and evolution is not good at navigating those.

example: depending on new CSI that comes along, monkey predecessor-eye might have evolved into a stalk-eye/snail eye, but it didn't. the fact that it didn't cannot reduce the probability of having evolved CSI reduced in your calculation. because you wanted to show that evolution is unlikely to produce any new CSI, not just the particular one you're talking about. You are not allowed to exclude that monkeys could have stalk-eyes from your calculation, or any other potential course of evolution from the probability calculation, because only then you are allowed to claim that it would be unprobable to gain any CSI.
You are misunderstanding the concept of CSI here and how it applies to ID, and you are again assuming evolution could evolve whatever object (real or possible). If the information change is measured above a threshold, it is CSI. If below the threshold, it is not CSI and ID would chalk the change up to a natural non-ID process. Monkey predecessor-eye -> monkey eye would have one measurement. Monkey predecessor-eye -> stalk eye would be another measurement. Monkey predecessor-eye -> *other* are other measurement(s). Even assuming each individual -> change would on it's own be a CSI event, yes, accounting for multiple possible change events would reduce the CSI calculation as a whole and reduce it's probability. But please realize you do not have an unlimited number of possible -> changes that are viable (empirically supported), and each instance of a viable possibility would reduce the CSI probability in an additive fashion and not multiplicative, and with the extremely low probability factors involved, it would probably not impact the CSI much at all.

ID would not exclude these other possibilities, but neither does it really care much. These are all just imagined instances of a process that has not been shown likely to be able to produce said non-existent objects.

in crystals ok. in evolution, you have NS - your perfect specificity selector, that chooses the right specificity at the right time and keeps evolving new useful traits. just not your specificity.
Other workable specifications, if known and beside the one in play, are considered in ID.

NS does not care about what you call a "higher function". as long as something is beneficial, it will be selected. the eye probably had the same main function all the time, not because the parts that the eye is made-of somehow support sight, but because sight is a far more useful feature of that thing, placed in all our predecessors, than any other possible use for the eye. But it's still possible that the very very first eye evolved from an ear or some antennae on the head of the premordial fish.
Yes, NS selects what is beneficial from whatever has changed. But you still have to explain the process of how the changed thing arrived. You cannot just assume it shows up, and bam! NS likes it. Especially when it looks like most new functions (up for selection) would require multiple changes (mutations) before NS could even begin to act. You have to account for those changes, and if you are relying on random changes to produce them, you are going to be waiting a long, long time before anything is produced for NS to work with.

the bacterial flagellum is a locomotion machine. it however has the same structure like a poison injector on other bacteria, and maybe evolved like a poison injector for a while untill a mutation made it movable at all, even slightly! and then NS took over and realized that locomotion was more beneficial use for this thing than poison injection.
This is laughable. The links between the Type 3 Secretion System, your "poison injector", and the bacterial flagellum are weak. The T3SS uses 10 proteins as a protein pumpt, and the bacterial flagellum uses 35 mostly different proteins to provide motion for a cell. Two totally different functions. The T3SS itself could be considered an irreducibly complex machine.

This co-option argument tries to account for one machine by saying it borrowed parts from another machine. But you still have not accounted for where the original parts come from to begin with. If you run the argument on, eventually you wind up borrowing from nothing. It also fails to account for the assembly instructions, which are even more complex than the parts. This argument is an equivocation.

Taking away parts of the flagellum destroys the ability of the system to act as a rotary propulsion machine, regardless whether a subset part of it could be used somewhere else to provide transport of proteins across a membrane. The flagellum is still irreducibly complex. What's more, the function of transporting proteins has as little directly to do with the function of rotary propulsion as a toothpick has to do with a mousetrap. So discovering the supportive function of transporting proteins tells us precisely nothing about how Darwinian processes might have put together a rotary propulsion machine.

basically the assumption of irreducable complexity is simply false. when you remove parts of the machine, the machine stops being useful for that one thing, that is true, but it's still useful for another.
Despite the video you included, you are wrong on this. In that program, David DeRosier, Kenneth Miller, and the Nova program in general gets the definition of IC wrong. It doesnt really matter if sub parts could be used elsewhere. Behe already acknowledged co-option as a possibility when he defined irreducible complexity. Other functions that might be assigned to parts have little or nothing to do with the function of the main system, like the flagellum. And saying that IC is false just because a part can have another function apart from the whole is like saying my laptop is not irreducibly complex just because I can use the hard drive as a doorstop.

the problem that evolution opponents have is that they assume one singleminded unalterable purpose for the trait. which is basically a circular argument they unknowingly commit into when arguing for ID, by assuming one single purpose and then deducing intelligence that is required to have that purpose .
Hardly. So, if ID looks at one object and determines CSI in it's creation, and assigns a probability of > 1 in 10^150, yet you come along and find another possible path, woohoo, now you have bettered your odds to 2 in 10^150. How many paths do you think you can come up with before you can make a dent in the probability.

#### WookieeB

##### Active Member
Evolution has been observed first hand, in almost (if not more by now since it was years ago that I sifted through them) one hundred separate instances utilizing various animals.

Give me an example of 2 instances of this, please. If you can include one of the "extra body parts" instances, that would be a bonus.

#### Teax

##### huh?
I think you are missing a point in your analogy. In your analogy, every object in existence has wheels.
(that is why in part I do not think it is a good analogy, it forces a person to think in a non-realistic manner to describe a realistic situation that is still under consideration. An analogy is supposed to simplify things or at least relate a subject to a more understandable experience)
you have been vastly influenced by side-points in our past discussion, I had to filter everything else out to point out where the flaw lies.

every analogy has flaws if you go far enough away from the subject at hand. the only perfect analogy for real life is real life.

For the illustration of my point, this analogy is more than adequate: I'm trying to explain why you cannot just use inductions everywhere you like.
the vehicle analogy does simplify, for step_three aka "positive evidence" of ID. the fact that you made the same bias mistake in the analogy as in the "positive evidence" argument tells me the analogy was good enough for this purpose

That is not a "4 wheels for sure" statement, but rather a "If it looks like a duck, smells like a duck, quacks like a duck, then its probably a duck" statement.
that statement does not change anything. you keep repeating this point but do you understand that that is implied in every induction?
y2b. for all big vehicles in this parking lot, can we infer that they probably have 4 wheels?
same thing as
y2b. for all big vehicles in this parking lot, can we infer that they have 4 wheels?
because the probably is automatically implied. Because it's an induction.

Yes, the wheel count of big vehicles is unknown. But the wheel count of small vehicles is known, and to our experience they ALL have 4 wheels. We now are presented with something (big vehicles) that bear many similar properties to small vehicles to which we know the wheel count. We know it has wheels, but we cannot see the wheel count on the big vehicles, but as we can relate much of it to small vehicles, and we have never seen a small vehicle with not 4 wheels, it is not a stretch to infer that big vehicles have 4 wheels too.
(...)
Up to the bold text you were doing fine. Without a theory of evolution, you can definitely induce something.
If in our experience, all vehicles where we can see the wheels are small vehicles, and the wheel count has always been 4, and we have never yet seen or heard of a real vehicle without 4 confirmed wheels. And now we see something that resembles a small vehicle, but that is just bigger, but it has it's wheels hidden. So, despite all our experience with wheels + vehicles, we cannot make an educated guess? We cannot link to our experience of small vehicles and only 4 wheels and infer that a similar object (but bigger) does not also have 4 wheels? Especially when we have been given no realistic reason to think that it may have other than 4?
no we can't.

I hope you read the explanation what a bias is, in the OP post, where I explain induction. (I added wiki links and examples), only then the following will make sence to you:

y1. all small vehicles in this parking lot, of which we know how many wheels they have, have 4 wheels.
y2. for all small vehicles in this parking lot, can we infer that they have 4 wheels
-- no-bias assertion: premise only has small vehicles, the conclusion only has small vehicles, seems fine
y2b. for all big vehicles in this parking lot, can we infer that they have 4 wheels
-- bias challenge: premise only contains small vehicles, conclusion contains big ones.

untill you refute the bias challenge by showing a reason why the wheel count of small vehicles should apply to big vehicles, the induction is biased and completely useless. you cannot just say "we relate much of it", how does what you relate bind the wheelcount?
the burden of proof in on your side, but you probably believed it wasn't because you said this:
"Especially when we have been given no realistic reason to think that it may have other than 4?"

It's nice to muse about what something potentially might have alternately developed into, but that is all just wishful thinking, made-up fantasy, there is no evidence of it existing.
It's nice to muse about other possible results of the random number generator, but it's all made-up fantasy, wishful thinking, there is no evidence of existing.

both statements are true, because we live in a universe where we have been given this particular random number/this particular development paths for each species.
this fact does not disprove that other paths were possible. you are not allowed to assume it based on lack of evidence, because there would be no evidence by-definition - which means lack of evidence is not evidence of anything.

Analogy of your argument: well the mailman took this route for delivery, so theres no possible other route he could have taken, because we have no evidence for it.
But in reality if the mailman took one route, he automatically, out of logical necessity, did not take another route. the fact that you have no evidence for another route does not make it reasonable to believe that no other route exists. The reality is we simply cannot know for sure, but we can examine the underlying system and make the statement: the mailman could have taken another path because he was free to choose and his vehicle can drive on any road. This statement is not based on evidence, but on logical necessity
, just like the alternate paths in evolution.
And even if you still insist on considering the open-ended object, you are assuming the very process that you are trying to explain, that is evolution.
I am doing exactly what you claimed to do: infer to the best explanation. since evolution has explanatory power here, and nothing else does, evolution is currently our best inference. There are no unreasonable assumptions involved.

Evolution is the open-ended option. Your argument is not. Thus you are refuting a straw man, that you constructed due to lack of understanding what evolution is. By showing that WookieeB-evolution ( ) cannot account for CSI you are not showing that evolution cannot account for CSI.

No assumptions are being made, but I am dealing with what we have, what is real.
But that is not true, you are assuming that the current evolutionary path is the only one that could have happened. that is an unreasonable assumption because it is not supported by anything, and your conclusions based upon this assumption are invalid aswell.

...but how did it appear? That is the question?!? You and Dawkins are doing the same thing and assuming what you are trying to prove. You are begging the question. It doesn't have to appear! You are only saying that because it exists. But evolution has no forethought, no plan, it doesnt have to do anything. You cannot assume the process just because the object you are trying to explain the cause of exists.
mutation.

Dawkins description of Mt Improbable is assuming a fitness landscape that is always uphill up toward whatever evolutionary property he is talking about (nautilus eye), and assuming that each peak jump is a minor probability event (like a single mutation). But that is not a realistic scenario for most functional changes. There would be valleys in between different fitness peaks, and evolution is not good at navigating those.
the 'landscape' is always changing. what is beneficial today might not be tomorrow. you again assumed one singleminded purpose, but NS is adaptable to the situation. In dawkings example a timeframe was implicitely chosen where the landscape didn't change significantly over the relevant period of time.

You are misunderstanding the concept of CSI here and how it applies to ID, and you are again assuming evolution could evolve whatever object (real or possible).

If the information change is measured above a threshold, it is CSI. If below the threshold, it is not CSI and ID would chalk the change up to a natural non-ID process. Monkey predecessor-eye -> monkey eye would have one measurement. Monkey predecessor-eye -> stalk eye would be another measurement. Monkey predecessor-eye -> *other* are other measurement(s). Even assuming each individual -> change would on it's own be a CSI event, yes, accounting for multiple possible change events would reduce the CSI calculation as a whole and reduce it's probability. But please realize you do not have an unlimited number of possible -> changes that are viable (empirically supported), and each instance of a viable possibility would reduce the CSI probability in an additive fashion and not multiplicative, and with the extremely low probability factors involved, it would probably not impact the CSI much at all.

ID would not exclude these other possibilities, but neither does it really care much. These are all just imagined instances of a process that has not been shown likely to be able to produce said non-existent objects.

Other workable specifications, if known and beside the one in play, are considered in ID.

Yes, NS selects what is beneficial from whatever has changed. But you still have to explain the process of how the changed thing arrived. You cannot just assume it shows up, and bam! NS likes it.
simple: Mutation. it shows up and bam! NS likes it.

Especially when it looks like most new functions (up for selection) would require multiple changes (mutations) before NS could even begin to act.
It is unreasonable to assume this. You even subconsciously realized this yourself because you just said 'looks like' und not 'is'.
A step by step improvement where only 1 trait/part is added to the machine on every CSI level is possible in every example you have given - and it will result in the final machine. the fact that the final machine has several parts does not help your argument.

You have to account for those changes, and if you are relying on random changes to produce them, you are going to be waiting a long, long time before anything is produced for NS to work with.
we did wait for a long time. just your calculation of how long we had to wait didn't account for an open-ended probability. It's still a long time thought, even with the open ended probability, as our own evolution tree shows.

This is laughable. The links between the Type 3 Secretion System, your "poison injector", and the bacterial flagellum are weak. The T3SS uses 10 proteins as a protein pumpt, and the bacterial flagellum uses 35 mostly different proteins to provide motion for a cell. Two totally different functions. The T3SS itself could be considered an irreducibly complex machine.

This co-option argument tries to account for one machine by saying it borrowed parts from another machine. But you still have not accounted for where the original parts come from to begin with. If you run the argument on, eventually you wind up borrowing from nothing. It also fails to account for the assembly instructions, which are even more complex than the parts. This argument is an equivocation.

(...)

Hardly. So, if ID looks at one object and determines CSI in it's creation, and assigns a probability of > 1 in 10^150, yet you come along and find another possible path, woohoo, now you have bettered your odds to 2 in 10^150. How many paths do you think you can come up with before you can make a dent in the probability.
this links with what I have said above about you assuming there are no other paths and that's an unreasonable assumption. there are no 1, 2 or 10 paths, there are an uncountable amount of paths. every mutation, no matter how positive and negative in relation to NS, changes the subset of the species. the possibilities are vast.

Despite the video you included, you are wrong on this. In that program, David DeRosier, Kenneth Miller, and the Nova program in general gets the definition of IC wrong. It doesnt really matter if sub parts could be used elsewhere. Behe already acknowledged co-option as a possibility when he defined irreducible complexity. Other functions that might be assigned to parts have little or nothing to do with the function of the main system, like the flagellum. And saying that IC is false just because a part can have another function apart from the whole is like saying my laptop is not irreducibly complex just because I can use the hard drive as a doorstop.
irreducable complexity in your definition does not matter for evolution. you make it sound like it's a point that somehow disproves evolution but i have not yet heared an actual argument why it does, despite your faulty probability statements. so please make it clearer

Taking away parts of the flagellum destroys the ability of the system to act as a rotary propulsion machine, regardless whether a subset part of it could be used somewhere else to provide transport of proteins across a membrane. The flagellum is still irreducibly complex. What's more, the function of transporting proteins has as little directly to do with the function of rotary propulsion as a toothpick has to do with a mousetrap. So discovering the supportive function of transporting proteins tells us precisely nothing about how Darwinian processes might have put together a rotary propulsion machine.
if you're fishing for an in-detail explanation of what enviroments influenced the NS to select what specific mutation, we might never find out. I fail to see why this matters thought, since you are not allowed to go backwards in evolution, only forwards when arguing.
you cannot say "if you take away a part", evolution works forwards, by making changes and then optimising so the new mutation is at maximum benefit. if you take away a part from the re-calibrated system, you obviously end up with a useless machine. but it was not useless before re-calibration to the new usage.

#### Jennywocky

##### guud languager
if you're fishing for an in-detail explanation of what enviroments influenced the NS to select what specific mutation, we might never find out. I fail to see why this matters thought, since you are not allowed to go backwards in evolution, only forwards when arguing.
you cannot say "if you take away a part", evolution works forwards, by making changes and then optimising so the new mutation is at maximum benefit. if you take away a part from the re-calibrated system, you obviously end up with a useless machine. but it was not useless before re-calibration to the new usage.

right -- pieces can and have developed independently but then over future iterations can change to become dependent on each other... they integrate, and the beneficial functionality will be selected for. At that stage, you can't remove one without the machine failing; but it doesn't mean they were integrated to begin with.

#### WookieeB

##### Active Member
http://scholar.google.com.au/scholar?q=evolution+speciation&btnG=&hl=en&as_sdt=0%2C5

Speciation in plants, insects, mammals, reptiles, birds. There's closer to 200 than 2.

Not wasting any more time on you ignoramus.

Yup, as I thought. You are good at doing a search, but you are apparently bad at reading.

You originally claimed "Evolution has been observed first hand...", and when pressed for examples you provide.... links. Did you even check what was in those links? I doubt it.

First one on the list talks about ants. Problem is, the writer didn't OBSERVE anything, he just is conjecturing about some polytypic meta-ant species that supposedly dispersed into multiple groups that then interbred amongst themselves and eventually lost breeding ability with other groups. Even if he had observed it (and he didnt) at best you just have a microevolutionary change, and most likely a loss of function at that. No big deal and nothing ID has a problem with.

2nd one on the list, again no observation, but conjecture on social evolution among animals. Story-telling that may be true, but there is no evidence for it yet. Oh, and by the way, the entire article is not available.

3rd, 4th, 5th on the list.. guess what, you can preview the first page, if even that, and there is no indication by the abstract that they are proposing any actual cases.

I stopped there, as I do not feel the need to go through the supposed 200 cases and do your work for you.

It is evident you have no idea what you are providing as proof, since your links lead to mostly unavailable information.

So again. Why dont YOU state 2 observed cases of evolution, something like the "growing entirely new body parts" you were so excited about. Or if you have problems with the concept of counting to 2, just provide me your one best case, hmm?

#### redbaron

##### irony based lifeform
The reason why I didn't is because like you said, there's so much research related to evolution that finding specific articles I read 3 years ago is difficult. Do you really think the extent of my research was googling that link? It wasn't. Thing is, I've already read the studies and I'm not interested in going on a hunt for them all over again.

Speciation has been observed. If you're too lazy and ignorant to find the information for yourself, I'm not going to spoonfeed it to you - especially when it's not even necessary to debunk your astoundingly ignorant statement, it's just icing on the cake. Eyeballs too complex for evolution...how desperate the ID proponents must be getting when that's the best they can come up with.

Since I'm such a nice guy though, here's a link to follow:

http://lmgtfy.com/?q=observed+instances+of+speciation

Good luck!

#### WookieeB

##### Active Member
The reason why I didn't is because like you said, there's so much research related to evolution that finding specific articles I read 3 years ago is difficult. Do you really think the extent of my research was googling that link? It wasn't. Thing is, I've already read the studies and I'm not interested in going on a hunt for them all over again.

So evolution is SOOOOO obvious that you would need to hunt back for something you read 3 years ago to answer my simple request.

Name me your best case of evolution.

Speciation has been observed. If you're too lazy and ignorant to find the information for yourself, I'm not going to spoonfeed it to you - especially when it's not even necessary to debunk your astoundingly ignorant statement, it's just icing on the cake. Eyeballs too complex for evolution...how desperate the ID proponents must be getting when that's the best they can come up with.

Funny thing is, a speciation event is not anything that would necessarily refute ID. ID allows for it, to an extent. So you are knocking down a strawman. Of course, for you it is not necessary to debunk my statement,.... probably because you cant. Sorry, bluster, assertion, and insults do not accomplish it.

The example of the eye is not the best example, but it is sufficient. Show how it could evolve.

Or...

Name me your best case of evolution.

Since I'm such a nice guy though, here's a link to follow:

http://lmgtfy.com/?q=observed+instances+of+speciation

Ahh, more links! I already said you are really good at coming up with those. No need to crow about it.

But sadly, links again don't support your assertion. Why don't you pick your favorite example from one of those links and we can discuss it.

or....

Name me your best case of evolution

Good luck!

Exactly!

#### redbaron

##### irony based lifeform
So evolution is SOOOOO obvious that you would need to hunt back for something you read 3 years ago to answer my simple request.

I actually already explained it without it. There's billions of years and quadrillions (if not more) of generational periods whereby evolution could cumulatively result in any of the complex biology we see today.e

There's no, "best case" of evolution. You've already accepted that evolution can be true on a micro scale. What you don't seem to realise is that all evolution is exactly the same thing. The only thing separating the evolution of something simple like bacteria and something complex like an eye - is time.

It's pretty clear that you either can't or don't want to understand this. The entire premise of saying that biological construct X can evolve but biological construct Y cannot does nothing but point to the fact that you have no idea what you're talking about.

Of course, for you it is not necessary to debunk my statement,.... probably because you cant.

Apart from the fact that your statement debunks itself, you haven't actually proven your statement, so there's nothing to debunk. All you've said is, "the eye is too complex to evolve".

~

The thing is, the theory of evolution is rather like the theory of gravity. Once you accept that evolution can happen - that's it. There's no arbitrary limit on the capacity for evolution to produce complexity. The only factors limiting it are generational periods and time. Of which there's been ample in the 3.5 billion years that life has existed on Earth.

Similarly you can't say that gravity exists, but black holes are impossible.

The only difference here is that evolution contradicts your chosen worldview that there exists a creator.

Which is why I initially skipped all this shit and got to the core of this whole argument in the first place:

You're a fucking idiot. If you weren't, this conversation wouldn't even be happening.

#### WookieeB

##### Active Member
There's billions of years and quadrillions (if not more) of generational periods whereby evolution could cumulatively result in any of the complex biology we see today.

Actually, no there is not. There are hard time limits. 3.5 billion years for the earth. that may sound like a lot to you, but it really is not enough time to produce a complex and SPECIFIED form like you think. The specification is what you are dodging from considering. The specification cannot, in even the long time scales you think are enough, cannot reasonably be generated by a random process, even if you have something like NS selecting. That is what the SCIENCE shows. You have not shown anything, you have just asserted that evolution can do something and then just assume that is the case. Where is your proof?

There's no, "best case" of evolution. You've already accepted that evolution can be true on a micro scale. What you don't seem to realise is that all evolution is exactly the same thing. The only thing separating the evolution of something simple like bacteria and something complex like an eye - is time.

Unfortunately this is not what the science shows, and that is why scientists that actually look into this stuff are abandoning neo-darwinism as the mechanism. Small changes by themselves are fine, doable, and proven to occur. Small changes coordinating together to develop a new function is not proven, not likely doable given the time scales we have. The coordination of micro changes is what is REQUIRED to develop the complext machines we see in life - that is the specification that you cannoot account for with RM + NS on any reasonable time scale. What ID is saying is that given what we know from science, the macro level machines found in life are not the result of micro steps, but are instead the result of specific coordinations of parts that are unlikely to come together by a random process within the hard time limits we have.

It's pretty clear that you either can't or don't want to understand this. The entire premise of saying that biological construct X can evolve but biological construct Y cannot does nothing but point to the fact that you have no idea what you're talking about.
It matters depending on what you are saying can evolve. changing the color of fur or something is easy, doable. Building a new part or function is not easy, or likely doable - because we can break down what is required to build the part or function, and then show mathematically that it is extremely unlikely. How unlikely? something to the tune of allowing a single person to win the Powerball lottery 18 times in a row and we wont question the odds, but if they win it 19 or more times in a row, then we start to question if something hinky is going on. You have to demonstrate that your micro steps can build a specified thing with better odds.

All you've said is, "the eye is too complex to evolve"
Nope. You are not paying attention, or more likely you are trying to ignore what I am actually saying: "The eye is too complex AND SPECIFIED to evolve". Deal with the specification and you might have a chance.

The thing is, the theory of evolution is rather like the theory of gravity. Once you accept that evolution can happen - that's it. There's no arbitrary limit on the capacity for evolution to produce complexity.

And here we have the money statement. Evolution can do anything. You have just admitted that evolution is unfalsifiable.

The only factors limiting it are generational periods and time.

Yup! ID is saying that.

Of which there's been ample in the 3.5 billion years that life has existed on Earth.

Nope. Unfortunately for you, the science is saying there has NOT been ample time.

The only difference here is that evolution contradicts your chosen worldview that there exists a creator.

Pot - Kettle

Which is why I initially skipped all this shit and got to the core of this whole argument in the first place:

You're a fucking idiot. If you weren't, this conversation wouldn't even be happening.

So you assert. But since you STILL have not presented a single real-life example of observed evolution (like creating a new body part), your credibility is waning, and your assertion means little.

#### redbaron

##### irony based lifeform
Well hey, this wasn't the experiment I was after but it seems that speciation is observed frequently enough that 10 minutes of google search revealed another one.

http://sandwalk.blogspot.com.au/2013/12/lenskis-long-term-evolution-experiment.html?m=1

Honestly the only credibility to lose here is with dipshits like yourself who seem to have an almost infinite capacity for misunderstanding and misinterpreting evolutionary theory.

Although I do really like how you've turned this argument away from your idiot comment about the eye being too complex to evolve naturally into somehow questioning my credibility. If anyone reading this really is gullible enough to fall for it, I don't want their credibility or respect.

You have just admitted that evolution is unfalsifiable.

No. Anyone with even an elementary understanding of evolutionary theory would know that there's plenty of ways to falsify it. The fact that you'd say something like this further highlights that you really have no idea what you're talking about when it comes to evolution.

You are not paying attention, or more likely you are trying to ignore what I am actually saying: "The eye is too complex AND SPECIFIED to evolve".

HAHAHA.

The fact you think that adding in the term specificity changes anything just makes your ignorance all the more blatant.

How unlikely? something to the tune of allowing a single person to win the Powerball lottery 18 times in a row

This must be one of the most hilariously stupid things I've ever read. But I'm going to let you prove your case. If you can do it, I'll even concede that you're right.

The chances of winning a lottery are 1/175,000,000

Now all you have to do is show everyone how you came to the conclusion that the evolution of the eyeball is less likely than winning the lottery not just once, but 18 times in a row. Must have taken a lot of time to calculate that sort of equation, with all of the variables involved. I'm sure it's something impressive and I'd love to see it.

I'd also be quite disappointed if you were so disingenuous that you made up these figures without actually having formulated any equations. That's quite intellectually dishonest and rather insulting to the intelligence of your audience that you'd assume they'd actually fall for that bullshit. You wouldn't do that though, right?

#### Teax

##### huh?
At this point I can't help but wonder if you genuinely have a mental defect or not. Just reading your posts makes me feel dumber.
That's a bit harsh. Theres nothing wrong with questioning the status quo.

~

Specificity is a nice idea for something else, but completely useless for calculating probabilities. As my last post showed, WookieeB does not yet fully grasp inductions, but those are also based on probability theory.

So the root cause seems to be the lack of understanding of probabilities. And if you assume one false thing, you can deduce anything from it, which explains the strong convictions. easy and without insults.

#### redbaron

##### irony based lifeform
There's also a difference between questioning the status quo and outright distorting scientific theory to further your own personal agenda.

#### WookieeB

##### Active Member
I've always liked magicians. There is an impressive skill in how they can misdirect and spotlight attention to another area while they do their hidden workings. The problem is though that if you watch their act enough, you start to not pay attention to where they are pointing and you may notice the intended hidden action. And once you know about it, it's hard to not then focus on the hidden actions. Unfortunately it kills the effect, because you realize how the "magic" really is just illusion.

You are a decent magician. But in the end we realize that all you are doing is pretending that something is real that in reality is not.

step_three in the OP post. it explains why step_three is an argument which by itself, is not a valid induction, and only as believable as step_two, if step_two were believable at all.

which references back to....

the "flaws" section:
step_three:....
pe_1: For all objects with CSI for which we found the cause, ID was always the cause
pe_2: For all objects with CSI, from pe_1 we infer(induce) that they have an intelligent cause.
bias challenge: the sample only contains non-life. because all the object, for which we found the cause, were non-living objects.
correction: For all non-living objects with CSI, from pe_1a we infer(induce) that they probably have an intelligent cause.
notice that we use the corrected version of pe_2 everyday.
Your bias challenge is junk, because I didnt bias out anything. The sample is looking at objects WITH CSI, and that includes lifeforms.

The whole question of the argument is to infer what is the origin of objects whose origins we cannot deduce. It would make no sense to include the property of the object in the sample as it is unknown, so you cannot include it.

But we are including it in the sample based on other criteria. I'll break it down to show this.

Lets ignore the origin property for the moment. So the sample will include objects with many other properties, which as a whole fall under an umbrella term of "CSI" or "not-CSI". We see those properties in some objects, and do not see them in other objects, but the sample includes ALL objects that we can measure the properties on, which obviously includes living and non-living objects. Lifeforms are in the sample. Within the entire sample, some objects do not have CSI, and others (including the lifeforms in the sample) do have CSI. There is no bias here, nothing has been selectively excluded.

So, now we want to look at the property of [origin] which can be only ID or non-ID. From the full sample (that includes lifeforms) that property is either filled in or not (known or unknown). For any object where we have a datapoint for [origin], we find that objects that have "CSI" are ALWAYS [origin]=ID, and objects that have "not-CSI" are either [origin]=ID or [origin]=non-ID. Lifeforms are among the total sample that do not have any datapoint for [origin], but you cannot say we are biased against lifeforms because there is no datapoint for [origin]. The whole point of the inference is to fill in something (provisionally) for the [origin] property. We are looking at the entire sample to server as the basis of that inference. And because we see that for property "CSI" that we have an entry for [origin] is ALWAYS [origin]=ID, then it is reasonable to infer that for objects with "CSI" would also be [origin]=ID. We have already explored all the other not-CSI objects, and that property is the only one to match to [origin]=not-ID objects.

So, there is no bias.

If you still insist that excluding the property that we are trying to determine is being a bias, then you also have no way to infer evolution. In fact you are admitting that you cannot infer anything where a property is unknown. And that would therefore invalidate any sciences like archaeology, forensics, cladistics, SETI and a bunch more. And it would also invalidate the scientific method, as per your stance on the argument you could not ever hypothesize about anything unknown.

The really illuminating thing is that it is YOU that is in fact showing a bias. You are setting up lifeforms as a special case where you cannot determine any specification. In effect you are saying that for lifeforms we cannot determine CSI.

#### WookieeB

##### Active Member
you have been vastly influenced by side-points in our past discussion, I had to filter everything else out to point out where the flaw lies.
every analogy has flaws if you go far enough away from the subject at hand. the only perfect analogy for real life is real life. For the illustration of my point, this analogy is more than adequate
So, because we cannot develop a "perfect" analogy, lets go ahead and use a bad one?

Your analogy is crap. And it doesnt even make sense logically with the premises you put in place.

vehicle
= object with CSI

small vehicle
= constructed objects/non-life

big vehicle
= lifeform

4 wheels
= intelligently designed

non 4 wheels/more then 4 wheels
= not intelligently designed

this parking lot
= things we observe that they at least exist. we can not necessarily observe everything about them, like their wheelcount. but we can see they're there.

knowing how many wheels a vehicle has
= inspected object and found it's origin. for exaple things you made, or saw how they are made. you know for sure they're ID

not knowing how many wheels a vehicle has
= object we did not analyse. like Mt. Rushmore, you don't analyze it to be 100% sure who made it, you just infer.

First, it is a bad idea to analogize a cause as an object. You run into all kinds of problems.
As it is, for your analogy, every item we can refer to as an object has to have wheels. Technically according to your design, if you want to identify the wheels of a vehicle apart from any other property of the vehicle, then your wheels are an object and they have to have wheels on themselves, so you have wheels on wheels. (and then those wheels have to have wheels on them, ad infinitum) So then how are you supposed to count how many wheels it has?

The only way around that is to make a vehicle a singular object itself, something that cannot be reduced in any objective way (which is effectively impossible in reality, so your analogy doesnt correspond to anything real). And if a vehicle is a singular object, then it is not possible for small vehicles to have 4 wheels, because it is the basest form it can be for its object type (similar to how an proton was viewed before we could smash them to pieces), and nobody is saying in the argument (even ID) that the base materials were ID. It also means that every small vehicle would have to be identical.

Your analogy is illogical.

the vehicle analogy does simplify, for step_three aka "positive evidence" of ID. the fact that you made the same bias mistake in the analogy as in the "positive evidence" argument tells me the analogy was good enough for this purpose
So if we ignore the illogical-ness of the analogy, and go at it from your viewpoint, then NO, I didnt use the same bias. See my previous post to cover this, but lets look at it from your analogy standpoint.

Vehicles exhibit CSI. So, we look at small vehicles - they all have a steering wheel, gearshift, windsheild, turn signal, headlights, seats..... and a 100 other properties that are unique to a vehicle. Now we look at big vehicles - they all have a steering wheel, gearshift, windsheild, turn signal, headlights, seats..... and the same 100 other properties that are unique to a vehicle. All vehicles (small and big) are included in our sample.

Your 6 wheeled fan-toy is also considered in the sample, but it does not have a steering wheel, gearshift,..etc, nothing that would tie it to being a "vehicle", so there would be nothing about it to tie to any inference.

ALL and EVERY small vehicle has 4 wheels. We cannot see how many wheels are on the big vehicle, but we know it has wheels. since it shares all the same properties of a small vehicle that we use to identify it as a vehicle, and as there are no vehicles known to be without 4 wheels, it makes perfect sense to infer the big vehicle also has 4 wheels.

I also noted your spoiler said: = "object we did not analyse. like Mt. Rushmore, you don't analyze it to be 100% sure who made it, you just infer." Which goes against your whole attempt to show me having a bias. Of course we dont analyze it 100%, because once you have identified it as specified complexity, you can logically infer.

no we can't.
... in relation to my statement of "relate much of it". Well, yes we can. CSI in small vehicles relates to the same CSI in big vehicles, all except the wheels portion, which is what we are trying to determine anyways and cannot inherantly be a part of the sample consideration since it is unknown.

hope you read the explanation what a bias is, in the OP post, where I explain induction. (I added wiki links and examples), only then the following will make sence to you:

y1. all small vehicles in this parking lot, of which we know how many wheels they have, have 4 wheels.
y2. for all small vehicles in this parking lot, can we infer that they have 4 wheels
-- no-bias assertion: premise only has small vehicles, the conclusion only has small vehicles, seems fine
Again, no bias, because it was included in the sample when we were checking for CSI properties.

untill you refute the bias challenge by showing a reason why the wheel count of small vehicles should apply to big vehicles, the induction is biased and completely useless.
I just did.

more to come later....

#### Teax

##### huh?
I've always liked magicians. There is an impressive skill in how they can misdirect and spotlight attention to another area while they do their hidden workings. The problem is though that if you watch their act enough, you start to not pay attention to where they are pointing and you may notice the intended hidden action. And once you know about it, it's hard to not then focus on the hidden actions. Unfortunately it kills the effect, because you realize how the "magic" really is just illusion.
that's exactly the way I see ID. an unfalsifiable statement supported by 3 flawed reasons to make it look falsifiable through misdirection to push it forward as a hypothesis, then an illusion of probability is presented (where there was none because the field is unknown) to make the other hypothesis look unlikely.

The whole question of the argument is to infer what is the origin of objects whose origins we cannot deduce. It would make no sense to include the property of the object in the sample as it is unknown, so you cannot include it. But we are including it in the sample based on other criteria.
yes! you're right: it would make no sence or to be more precise it would be fallacious. Hence the sample can not contain those objects, thus the orange statement illustrates that you misunderstood inductions/probability theory.

Your bias challenge is junk, because I didnt bias out anything. The sample is looking at objects WITH CSI, and that includes lifeforms.
(....)
I'll break it down to show this.

Lets ignore the origin property for the moment. So the sample will include objects with many other properties, which as a whole fall under an umbrella term of "CSI" or "not-CSI". We see those properties in some objects, and do not see them in other objects, but the sample includes ALL objects that we can measure the properties on, which obviously includes living and non-living objects. Lifeforms are in the sample. Within the entire sample, some objects do not have CSI, and others (including the lifeforms in the sample) do have CSI. There is no bias here, nothing has been selectively excluded.
a sample must be selected based on what population you are planning to make predictions about. The sample must contain objects where the relevant property [origin] is not just measurable, but actually measured. =It must be known. Your sample includes objects with an unknown cause, hence that sample is not suitable to infer a cause.

So, now we want to look at the property of [origin] which can be only ID or non-ID. From the full sample (that includes lifeforms) that property is either filled in or not (known or unknown). For any object where we have a datapoint for [origin], we find that objects that have "CSI" are ALWAYS [origin]=ID, and objects that have "not-CSI" are either [origin]=ID or [origin]=non-ID. Lifeforms are among the total sample that do not have any datapoint for [origin], but you cannot say we are biased against lifeforms because there is no datapoint for [origin].
If [origin] is unknown the object is not allowed to be part of the sample in the first place.

thus I can (must, actually) say that the sample is biased against lifeforms because there is no datapoint for [origin].

this guy even touches the fine-tuning argument, nice video:
http://www.youtube.com/watch?v=5SZhbEIBDMk

analogy :
there are red and green apples in the basket. but only some red apples are known to be sweet, we have no idea about the green ones, because you never tasted one.

trying to induce the sweetness: you cannot place the green apples into the sample just because they're apples. their sweetness is unknown. a sample for sweetness is by definition = all(or subset) apples that you tasted. you cannot induce the sweetness of green apples from the red apples that you tasted. red and green apples are 2 different types of objects. unless you additionally show why the sweetness of red apples is the same as in green apples, for example by somehow showing that the color of the apple is independent from its nutrients. but induction alone is invalid.

apple = object with CSI
sweet = intelligently designed
red = non-life
green = life
pomegranate = object without CSI

what does independence mean? here:
independent variables

The really illuminating thing is that it is YOU that is in fact showing a bias. You are setting up lifeforms as a special case where you cannot determine any specification. In effect you are saying that for lifeforms we cannot determine CSI.
setting lifeforms as a special case is as necessary as setting green apples in that analogy as a special case. why would CSI in life have the same cause as CSI in non-life? -> That's what independence of the properties [life] and [origin] means.
• Without step_two, step_three does not answer that question.
• With step_two, step_three is only as believable as step_two.
in both cases step_three aka "positive evidence" is pointless. forget about it.

If you still insist that excluding the property that we are trying to determine is being a bias, then you also have no way to infer evolution. In fact you are admitting that you cannot infer anything where a property is unknown. And that would therefore invalidate any sciences like archaeology, forensics, cladistics, SETI and a bunch more. And it would also invalidate the scientific method, as per your stance on the argument you could not ever hypothesize about anything unknown.
lol, you just basically said "if you still insist that a circular argument is a fallacy".... call me old fashioned, but yes, I do

when inferring the unknown induction is reasonable, Faulty Generalisation is not. the difference lies in trying to infer a property A across different properties of B, without showing that B is independent from A.

You are allowed from the few red apples you tasted to infer that all red ones are sweet, not the green ones. A = sweetness, B = color.

So, because we cannot develop a "perfect" analogy, lets go ahead and use a bad one?
EDIT: ok I get it, you don't like my analogy. I'll drop it.. there's no need for it anymore, since we're discussing the positive evidence directly.

First, it is a bad idea to analogize a cause as an object. You run into all kinds of problems.
As it is, for your analogy, every item we can refer to as an object has to have wheels. Technically according to your design, if you want to identify the wheels of a vehicle apart from any other property of the vehicle, then your wheels are an object and they have to have wheels on themselves, so you have wheels on wheels. (and then those wheels have to have wheels on them, ad infinitum) So then how are you supposed to count how many wheels it has?
funny! but it ties in with my previous statement:

every analogy has flaws if you go far enough away from the subject at hand. the only perfect analogy for real life is real life.
so youre basically hatin' on all analogies like ever... ok whatever... are you sure you're not an ISTP?

Vehicles exhibit CSI. So, we look at small vehicles - they all have a steering wheel, gearshift, windsheild, turn signal, headlights, seats..... and a 100 other properties that are unique to a vehicle. Now we look at big vehicles - they all have a steering wheel, gearshift, windsheild, turn signal, headlights, seats..... and the same 100 other properties that are unique to a vehicle. All vehicles (small and big) are included in our sample.

Your 6 wheeled fan-toy is also considered in the sample, but it does not have a steering wheel, gearshift,..etc, nothing that would tie it to being a "vehicle", so there would be nothing about it to tie to any inference.

I only need to find 1 property that you completely didn't include in your sample, and can exclude that property in the induction conclusion. That's how induction works, sorry...

the 6 wheel toy is not part of the sample because we were trying to infer something about vehicles.

CSI in small vehicles relates to the same CSI in big vehicles, all except the wheels portion, which is what we are trying to determine anyways
yep, you just refuted yourself. induction alone cannot determine the wheels. and you just admited that you cannot demonstrate statistical independence, to support the base induction.

I also noted your spoiler said: = "object we did not analyse. like Mt. Rushmore, you don't analyze it to be 100% sure who made it, you just infer." Which goes against your whole attempt to show me having a bias. Of course we dont analyze it 100%, because once you have identified it as specified complexity, you can logically infer.
Mt. Rushmore is not living. you do realize that right? that's why I could infer it's intelligent design. induction is valid there

Last edited:

#### WookieeB

##### Active Member
that's exactly the way I see ID. an unfalsifiable statement supported by 3 flawed reasons to make it look falsifiable through misdirection to push it forward as a hypothesis, then an illusion of probability is presented (where there was none because the field is unknown) to make the other hypothesis look unlikely.

If you cannot see how ID is falsifiable, you are not paying attention. Probability is not an illusion, and it can be demonstrated. Your problem isnt with the structure of the ID argument, though you are trying to say so via obfuscation. Your problem is in whether we can actually assign a probability to a complex specificaion, cause you are denying there is any specification.

Your problem is that you are just rejecting a hypothesis because you do not like it. You won't look at any evidence for an alternate hypothesis because in your mind there is no alternative evidence.

yes! you're right: it would make no sence or to be more precise it would be fallacious. Hence the sample can not contain those objects, thus the orange statement illustrates that you misunderstood inductions/probability theory.
You are being dishonest or you either misunderstood the statement. I'll grant you the latter.
"It would make no sense to include the property of the object in the sample as it is unknown" - this means it makes no sense to include the an object (lifeform) in the sample based solely on the property of [origin]. We agree here.

"But we are including it in the sample based on other criteria." - but we are including an object (lifeform) based on other criteria (other properties which happen be CSI). The "it" is not the property of [origin], which is not clear in just my orange statement, but should have been understood from the context of the rest.

The sample includes objects based upon the property of CSI - which includes all objects without it and all with it (including lifeforms)

a sample must be selected based on what population you are planning to make predictions about. The sample must contain objects where the relevant property [origin] is not just measurable, but actually measured. =It must be known.
(population you are planning to make predictions about) = all objects with CSI and without CSI, which basically includes every object that could be considered
(The sample must contain objects where the relevant property [origin] is not just measurable, but actually measured. = It must be known) = yes. CSI is the measurable property and we know the measurement.

Your sample includes objects with an unknown cause, hence that sample is not suitable to infer a cause.
And here you fall off the rails. The reason to make an inference is to explain an unknown property. If you really believe your statement, then you are saying that you cannot make an inference,.... ever, about anything.
If [origin] is unknown the object is not allowed to be part of the sample in the first place.

thus I can (must, actually) say that the sample is biased against lifeforms because there is no datapoint for [origin].
BS. This stupid.

Again, the sample is gathered based on CSI.

Since you like analogies so much, why dont you give me a clear example of an inference where the property under consideration is not known.

this guy even touches the fine-tuning argument, nice video:
http://www.youtube.com/watch?v=5SZhbEIBDMk
Nice video, but it made no point. It was just defining things.
analogy :
there are red and green apples in the basket. but only some red apples are known to be sweet, we have no idea about the green ones, because you never tasted one.
Analogizing ID, it would never make any inference on the sweetness of green apples, especially after you say "but only some red apples are known to be sweet" Here you have a "bias" against some apples.

what does independence mean? here:
independent variables

setting lifeforms as a special case is as necessary as setting green apples in that analogy as a special case. why would CSI in life have the same cause as CSI in non-life? -> That's what independence of the properties [life] and [origin] means.
• Without step_two, step_three does not answer that question.
• With step_two, step_three is only as believable as step_two.
in both cases step_three aka "positive evidence" is pointless. forget about it.

when inferring the unknown induction is reasonable, Faulty Generalisation is not. the difference lies in trying to infer a property A across different properties of B, without showing that B is independent from A.
Nice video, but it again made no point. Just defining independence, which is fine.

And here we have you finally acknowledging what I have been saying -

So, if we are talking about the properties of objects (not any inference yet, but just setting the premises) regarding the property of known origins, the two steps are.....NOT independent.
CSI property possibilities - noCSI | yesCSI | ? (unknown/not measurable)
ORIGIN property possibilities - non-ID | ID | ? (unknown)
step_two is the negative property for ID - 100% of known ORIGIN=non-ID does not produce [CSI=yesCSI]
step_three is the positive property for ID - 100% known ORIGIN=ID does produce [CSI=yesCSI]
You cannot separate the two steps, because it is making an observation regarding objects based on the same property [ORIGIN] where ORIGIN ≠ ?.

But the property of CSI is independent of the property of ORIGIN. So far ID is not saying that property yesCSI = originID, because it is not making any statement for items of unknown origin, so we can also have yesCSI = origin?.

The sample size is including every object where we can measure CSI, there no objects where CSI = ? The sample is including lifeforms, which have CSI, it is measurable and established. Lifeforms also have a property of ORIGIN, but that is independent of the property of CSI. For now, the lifeform's property of origin is "?". Lifeforms (and any other yesCSI objects of unknown origin) are a special or independent case right now, we have not yet made any premise statement regarding them, and they do not fall within the observations of properties as seen in step_two or step_three

Now we come to the induction, "when inferring the unknown induction is reasonable,".
Setting the stage -
What is the unknown? The ORIGIN property of some of the objects in our sample, a property that is independent of other properties of the objects.
What is the correlation property between ALL the objects that we will look at? property CSI.
We care about objects where the CSI property = yesCSI (which includes the objects were origin=?, which a subset is lifeforms). We don't care for any objects where the CSI property = noCSI, so they are excluded.
Known = we have evidence/data
unknown = we have no evidence/data

So, based on the observed correlation between two independent properties of objects in our sample, when such properties are known (
[step_two - 100% of known ORIGIN=non-ID does not produce [CSI=yesCSI]
AND
step_three - 100% known ORIGIN=ID does produce [CSI=yesCSI] ]
or in other words,
when we have evidence/data that an object has yesCSI AND we have evidence/data of the ORIGIN, in 100% of cases the ORIGIN=ID
or to simplify further,
if yesCSI, then when ORIGIN ≠ ?, ORIGIN=ID
)
we can infer, or evidence suggests, that when we see yesCSI, the ORIGIN=ID.

The only alternative to the inference is a case where an ORIGIN=non-ID ANDyesCSI, but there is no evidence/data to suggest this, so this is an imaginary (made-up) case.

So, I addressed your criteria and the argument is sound.

lol, you just basically said "if you still insist that a circular argument is a fallacy".... call me old fashioned, but yes, I do
But I didnt say that, because there is no circular argument.

You are allowed from the few red apples you tasted to infer that all red ones are sweet, not the green ones. A = sweetness, B = color.
But this statement is not analogous to ID, so it is irrelevant. ID would not say all red ones are sweet, since your premise said "but only some red apples are known to be sweet, ", duh! The analogy is making an inference on green apples, not red apples. If there is no established property of red apples, then we for sure wouldnt make any inference for the green.
Bad analogy again.

so youre basically hatin' on all analogies like ever... ok whatever... are you sure you're not an ISTP?
No, Im just hatin' on bad analogies. There is such a thing as a good analogy, even if it is not perfect.

I only need to find 1 property that you completely didn't include in your sample, and can exclude that property in the induction conclusion. That's how induction works, sorry...
which contradicts your prior statement: "when inferring the unknown induction is reasonable,". So maybe you do not know how induction works.

yep, you just refuted yourself. induction alone cannot determine the wheels. and you just admited that you cannot demonstrate statistical independence, to support the base induction.
No, not really. I said: "CSI in small vehicles relates to the same CSI in big vehicles, all except the wheels portion, which is what we are trying to determine anyways". I mistated the meaning, but you should be smart enough to understand. I should not have included "all except the wheels portion," because I really can't. The wheels property has nothing to do with CSI or any other physical property of vehicles, according to your analogies illogical rules. Wheels is statistically independent from CSI. The induction is to try to make a reasonable determination for wheels.

So I should have said:
"CSI in small vehicles relates to the same CSI in big vehicles. They share all the properties that make both small and big "vehicles". So in the absence of any data to suggest that a vehicle doesnt have 4 wheels, in our experience we can infer that a vehicle has 4 wheels, which is what we are trying to determine anyways"

Mt. Rushmore is not living. you do realize that right? that's why I could infer it's intelligent design. induction is valid there
[/spoiler]
And here you actually make MY point.

For the ID argument, MT Rushmore is in the same category as lifeforms. Both are an object(s) whose origin is unknown but who have CSI. In that manner there is no distinction. I include them both - no bias. So, if you can make a valid induction for one CSI object of unknown origins, then you should be able to make a valid induction for another CSI object of unknown origins.

You are the one with the bias. You are the one excluding lifeforms for some unstated reason. But we both know what that reason is.... you do not WANT lifeforms to be included.

So once you admit that the inductive argument structure for ID is valid, we can move on to discussing the areas that everyone else argues - 1) life does not contain CSI, or 2) non-ID processes can produce CSI.

#### Teax

##### huh?
I hope your browser supports these characters, if not try to zoom in/out:
Code:
(─) (│) (░) (█) (▓) (►) <-- all brackets have to be filled

[img]http://intpforum.com/attachment.php?attachmentid=2391&stc=1&d=1415616693[/img] <- should look something like this screenshot
-------------------------------------------------------------------------------
If you cannot see how ID is falsifiable, you are not paying attention.
(...)
Your problem is that you are just rejecting a hypothesis because you do not like it. You won't look at any evidence for an alternate hypothesis because in your mind there is no alternative evidence.
(...)
You are the one with the bias. You are the one excluding lifeforms for some unstated reason. But we both know what that reason is.... you do not WANT lifeforms to be included.
(...)
So once you admit that the inductive argument structure for ID is valid. we can move on to discussing the areas that everyone else argues - 1) life does not contain CSI, or 2) non-ID processes can produce CSI.
sample bias has nothing todo with people bias though, I wasn't calling you biased. not that explicitely anyway
i do like it. I don't like ID because it's fallacious. I might share my views on who's the intelligent designer if this debate is over in our lifetimes

Probability is not an illusion, and it can be demonstrated. Your problem isnt with the structure of the ID argument, though you are trying to say so via obfuscation. Your problem is in whether we can actually assign a probability to a complex specificaion, cause you are denying there is any specification.
here's a copypaste from OP: Specified = necessary to function = if you take it out, the whole system becomes useless/stops working. irreducable complexity.
• I concede that specificity is definable.
• I concede that specificity is measurable (just take away one part and see what happens).
• I concede that lifeforms have specificity.
my claim: specificity is irrelevant. I heared you bring 2 arguments against evolution. please present these 2 points that I'm only vaguely stating:
• specificity, the fact that a machine is useless without a part. please state, as clear and concise as possible, why this is an argument against evolution. I will then attach my claim as a response. (so I don't attack a strawman)
.
• probability to produce CSI by evolution is too low to be believable. redbaron challenged you here, don't leave him hangin' .

The sample must contain objects where the relevant property [origin] is not just measurable, but actually measured. = It must be known) = yes. CSI is the measurable property and we know the measurement.
see the difference?

You are being dishonest or you either misunderstood the statement.
(...)
"But we are including it in the sample based on other criteria." - but we are including an object (lifeform) based on other criteria (other properties which happen be CSI). The "it" is not the property of [origin], which is not clear in just my orange statement, but should have been understood from the context of the rest.
yes I undertood that correctly. I don't even know where you could have gotten this from... I thought you merely misunderstood induction, but you're the first one I met to misunderstand what a sample is. youtube it. google it. and here:

definition:
sample = premise of an inductive-inference. like picking people on the street, ask them about X, to infer something about X of the whole country.

randomly pick a couple of men in your city and ask them if they like fast-food. if all of them do, you can induce that all men in that city like fast-food.

for ID, X = [origin]

Code:
┌──────────────────────────────┐
│██ Population ████████████████│
│██ unknown X  ████████████████│
│██████████████████████████████│
│██████████████░ Sample ░░░░███│
│██████████████░ known X░░░░███│
│██████████████░░░░░░░░░░░░░███│
│██████████████████████████████│
└──────────────────────────────┘
the objects with unknown X are the ones you are making the conclusion about. thus a sample by definition cannot contain objects with unknown X.
when you include objects that you are trying to make a conclusion about = objects with unknown X into the premise(=sample), it's called: circular reasoning, because they are part of the conclusion.

Your sample includes objects with an unknown cause, hence that sample is not suitable to infer a cause.
And here you fall off the rails. The reason to make an inference is to explain an unknown property. If you really believe your statement, then you are saying that you cannot make an inference,.... ever, about anything.
fascinating statement. I'll give you some rope: how did you deduce that? please elaborate!

which contradicts your prior statement: "when inferring the unknown induction is reasonable,". So maybe you do not know how induction works.
you really believe induction is suitable to infer any unknown? induction is just the opposite of deduction, it's not a magic wand. it has limits

basic induction:
Code:
▓ = has X
sample = premise                    induction = inference that the space
premise = a couple objects          between the points is same as the points
┌───────────────────────────────┐  ┌───────────────────────────────┐
│▓                      ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓           ▓             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│         ▓                     │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│                      ▓     ▓  │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              ▓                │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│▲    ▓                 ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││    ▲         ▓             ▓ │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▲▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││ ▓  │                         │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└┼────┼─────────────────────────┘  └───────────────┼───────────────┘
│    │                                            │
│    that's an object that we know has X          │
the empty space are objects with unknown X        │
│
the inference tells us that every object probably has X

at any point in the future, any air-head can come along and point out: hey, all your objects are RED!!! ~~~*girly scream*

you have to take him seriously. if you just stick your fingers into your ears and cry 'lalala' you would be cherry picking
once a bias in the already existing sample(=premise) has been discovered, the conclusion only applies to the same sub-population.

Code:
premise=sample:                    population=conclusion:
▓ = has X
red:            green:              red:            green:
┌───────────────┬───────────────┐  ┌───────────────┬───────────────┐
│▓       ▓      │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│  ▓    ▓       │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│          ▓    │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│        ▓   ▓  │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│   unknown     │
│        ▓     ▓│               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│  ▓  ▓   ▓     │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│  ▲        ▓   │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│  │            │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
└──┼────┬───────┴──────┬────────┘  └───────────────┴───────────────┘
│    │              │
│    │         objects with property [color=GREEN]
│ this are all objects with property [color=RED]
this objects has a known X additionally to having [color=RED]
bias means that the sample(=objects with known X) does not contain objects with the additional property , it's a bias against this color.

I only need to find 1 property or property-combination that you completely didn't include in your sample, and we must completely exclude that property in the induction conclusion. That's how induction works, sorry...

And here you actually make MY point.

For the ID argument, MT Rushmore is in the same category as lifeforms. Both are an object(s) whose origin is unknown but who have CSI. In that manner there is no distinction. I include them both - no bias. So, if you can make a valid induction for one CSI object of unknown origins, then you should be able to make a valid induction for another CSI object of unknown origins.
No. by discovering that mt. rushmore is not alive, and lifeforms are, they've been automatically assigned different categories. it's what's called "new evidence". you cannot stick your fingers in your ears and ignore it, otherwise you're pleading special case. look >here<, see how they split the table into rows and columns? every variable combination is being considered.
true.
I assume you mean in the sample. you cannot because you yourself admitted the origin is unknown. putting those objects part of the sample would be same as pretending like the origin is known => circular argument. since the origin is what we're trying to infer.
that is a nonsequitor. true - the 2 objects have some common properties, but that is irrelevant. we have found a difference. let's pretend we're one of the smart guys and not ignore variables.

by assuming that this difference is irrelevant for your induction, you are implicitely assuming that there is no difference!!! -> which is a statement that we can not know if it's true, theres no evidence (see below schematic)

So, if we are talking about the properties of objects (not any inference yet, but just setting the premises) regarding the property of known origins, the two steps are.....NOT independent.
*sigh* ok, no matter. superposition of arguments is not the best explanation tool, my bad for choosing it

CSI property possibilities - noCSI | yesCSI | ? (unknown/not measurable)
ORIGIN property possibilities - non-ID | ID | ? (unknown)
step_two is the negative property for ID - 100% of known ORIGIN=non-ID does not produce [CSI=yesCSI]
step_three is the positive property for ID - 100% known ORIGIN=ID does produce [CSI=yesCSI]
You cannot separate the two steps, because it is making an observation regarding objects based on the same property [ORIGIN] where ORIGIN ≠ ?.

But the property of CSI is independent of the property of ORIGIN. So far ID is not saying that property yesCSI = originID, because it is not making any statement for items of unknown origin, so we can also have yesCSI = origin?.

The sample size is including every object where we can measure CSI, there no objects where CSI = ? The sample is including lifeforms, which have CSI, it is measurable and established. Lifeforms also have a property of ORIGIN, but that is independent of the property of CSI. For now, the lifeform's property of origin is "?". Lifeforms (and any other yesCSI objects of unknown origin) are a special or independent case right now, we have not yet made any premise statement regarding them, and they do not fall within the observations of properties as seen in step_two or step_three

Now we come to the induction, "when inferring the unknown induction is reasonable,".
Setting the stage -
What is the unknown? The ORIGIN property of some of the objects in our sample, a property that is independent of other properties of the objects.
What is the correlation property between ALL the objects that we will look at? property CSI.
We care about objects where the CSI property = yesCSI (which includes the objects were origin=?, which a subset is lifeforms). We don't care for any objects where the CSI property = noCSI, so they are excluded.
Known = we have evidence/data
unknown = we have no evidence/data

So, based on the observed correlation between two independent properties of objects in our sample, when such properties are known (
[step_two - 100% of known ORIGIN=non-ID does not produce [CSI=yesCSI]
AND
step_three - 100% known ORIGIN=ID does produce [CSI=yesCSI] ]
or in other words,
when we have evidence/data that an object has yesCSI AND we have evidence/data of the ORIGIN, in 100% of cases the ORIGIN=ID
or to simplify further,
if yesCSI, then when ORIGIN ≠ ?, ORIGIN=ID
)
we can infer, or evidence suggests, that when we see yesCSI, the ORIGIN=ID.
ugh.... I will do you the courtesy to ignore the orange statements since your definition of a sample is still something that is neither commonly accepted nor understood by me.

I suspect you misunderstood what independence means. it's mutually exclusive with correlation, look it up: correlation, independance.

awesome, you see the correlation between CSI and [origin], let me use the same reasoning: how do you disprove correlation between [life] and [origin]?

hypothetically: if living CSI did have a natural cause, [origin] would correlate to both, CSI and [life], at the same time.

it logically follows: to show ID, you have to show that [origin] correlates to CSI (done), but not to [life].
=> show that [life] is statistically independent from [origin]
=> no matter if it's life or not, origin must stay the same
=> life/notlife doesn't have any implication upon the origin
=> [origin] does not correlate with [life]
=> a statement that we can not know if it's true. It's begging the question.

technically it's a form of circular reasoning, because if you force the inference, you are essentially assuming the conclusion and deducing the "begged question" from the conclusion.

Code:
negative evidence           ████████████████
¬life + ¬CSI ~ ¬ID  ──────► ██ several    ██ ──────►  CSI => ID
¬life +  CSI ~  ID  ──────► ██ inductions ██
positive evidence           ████████████████              [color=Olive]│[/color]
▲                          [color=Olive]│[/color]
independance [life]<->[origin] │                          [color=Olive]▼[/color]
required│                    [color=Olive]life + CSI => ID[/color]
you didn't show, thus implicit│                          [color=Olive]│[/color]
│                          [color=Olive]│[/color]
┌──────────────────────────────┐        [color=Olive]can only be│deduced this way[/color]
│no matter if it's life or not,│[color=Olive]◄──────────────────┘[/color]
│origin will stay the same     │     [color="DarkGray"]confirmation-biased people are[/color]
└──────────────────────────────┴─────[color="DarkGray"]more likely to just assume[/color]
▲            [color="DarkGray"]this convenient thing[/color]
│
│
[color=Pink]missing evidence[/color]
the whole construct is only useful if you want to mislead people~~~ *surprise* if you want to conceal that you're missing a key piece of evidence. all of this jazz is just one big empty argument, one giant waste of time.

let's see how you weasel yourself out of this!

here's what the ID argument looks like you presented it.
Code:
░ = object with natural cause
▓ = object with ID
sample:                          conclusion:
¬CSI           CSI              ¬CSI           CSI
┌─────────────┬──────────────┐  ┌─────────────┬──────────────┐
│░     ░      │  ▓        ▓  │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ░  ░       │      ▓       │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│    ▓   ░    │              │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│             │   ▓ ▓        │  │░░░░░░░░░░░░▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ▓   ░     ░│           ▓  │  │░░░░░░░░░░░▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│    ░  ░     │              │  │░░░░░░░░░░▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓   ░   │      ▓       │  │░░░░░ partly │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ░          │  ▓    ▓  ▓   │  │░░░░░░░░▓▓▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└─────────────┴──────────────┘  └─────────────┴──────────────┘
back then it was ok to believe. I don't blame you. but that's the fun about inductions, they're only valid untill you are presented with bias in your sample. just like any statistical inference.

after dr. killjoy points out the bias, you have to admit that
Code:
░ = object with natural cause
▓ = object with ID
sample:                               conclusion:
¬CSI            CSI           ¬CSI            CSI
┌─────────────┬─────────────┐  ┌─────────────┬─────────────┐
│░     ░    ▓ │      ▓   ▓  │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓│
│             │   ▓   ▓     │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓│
¬life │     ░  ░ ░  │             │  │░░░░░░░░░░░░▓│▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ▓      ▓   │     ▓     ▓ │  │░░░░░░░░░░░▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓│
│    ▓ ░     ░│ ▓      ▓    │  │░░░░░░░░░░▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓│
├─────────────┼─────────────┤  ├─────────────┼─────────────┤
│             │             │  │             │             │
│             │             │  │             │             │
life │             │             │  │   unknown   │   unknown   │
│             │             │  │             │             │
│             │             │  │             │             │
└─────────────┴─────────────┘  └─────────────┴─────────────┘

But this statement is not analogous to ID, so it is irrelevant. ID would not say all red ones are sweet, since your premise said "but only some red apples are known to be sweet, ", duh! The analogy is making an inference on green apples, not red apples. If there is no established property of red apples, then we for sure wouldnt make any inference for the green.
Bad analogy again.
apple = object with CSI
sweet = intelligently designed
red = non-life
green = life
something like... lemon = object without CSI (assume its green)

ID does make an inference about the green apples, simply because they're apples and not lemons.
Code:
ID:
▓ = sweet
apple            apple             apple            apple
┌─────────────────────────────┐  ┌─────────────────────────────┐
│▓       ▓          ▓         │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ▓    ▓                     │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│          ▓             ▓    │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│        ▓   ▓                │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│        ▓            ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓   ▓        ▓          │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│           ▓                 │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ▓                       ▓  │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└─────────────────────────────┘  └─────────────────────────────┘

but once we have detected that the sample is fully biased towards the color red - theres no turning back:

Code:
sanity:
▓ = sweet
red + apple      green + apple     red + apple     green + apple
┌───────────────┬───────────────┐  ┌───────────────┬───────────────┐
│▓       ▓      │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│  ▓    ▓       │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│          ▓    │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│        ▓   ▓  │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│   unknown     │
│        ▓     ▓│               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│     ▓   ▓     │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│           ▓   │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│  ▓            │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
└───────────────┴───────────────┘  └───────────────┴───────────────┘

reflects the fallacious part of ID wonderfully and deliciously.

No, Im just hatin' on bad analogies. There is such a thing as a good analogy, even if it is not perfect.
got any good ones? with pictures plz, always fun. I'm actually only in here for the entertainment...

#### Attachments

• asdf.png
575 bytes · Views: 243

#### WookieeB

##### Active Member
I hope your browser supports these characters, if not try to zoom in/out:
Code:
(─) (│) (░) (█) (▓) (►)
-------------------------------------------------------------------------------

I would guess not, cause a lot of these pictures do not make sense to me. Perhaps if you could label them more or explain the graphical representations.

So, before I comment further, if possible, please answer some questions.

In the code boxes, I see a box with a white outline border. Is this representing the sample space? Does this apply both where you have in part a label of "sample" or "conclusion"?

Within these boxes, where you have small rectangular icons, what does a blank space represent? I'm guessing an unknown object or property But Im not sure which property is unknown. For instance, you have a Box labeled "CSI" and solid rectangles in it that are defined "objects with CSI", but what is a blank area in that box supposed to mean? Is it unknown if it has CSI or not, unknown for origin, or both?

And depending on the above clarification, you have outlined areas (often labeled with "conclusion") that are filled in. Would such a filled in conclusion allow for any unknown properties still?

A lot of your box diagrams do not list what the rectangular icons are supposed to represent, so I'm not sure about the point there. I'd rather not guess, so can you please clarify?

With respects to the Mt Rushmore example, by what criteria are you able to make any inference (since you said you could)?

The diagram under your text "after dr. killjoy points out the bias, you have to admit that" is suggesting that any ID inference is not possible. Then how can you infer something like evolution, or are you? (per the diagram, no inference seems possible)

Lastly, can you please expound or dumb down for me on how you think ID is making a circular argument? Again, the diagram is not understandable to me because the comments and labeling are not clear.

Thanks.

#### Teax

##### huh?
In the code boxes, I see a box with a white outline border. Is this representing the sample space?
the word "sample space" has special meaning in statistics(=possible values for X) that is a bit counterintuitive... the code boxes represent "all objects" or "object space". divided code boxes, contain all objects in total, but divide them among the boxes based on properties.

Does this apply both where you have in part a label of "sample" or "conclusion"?
yep, it's "before" and "after" the induction. before the induction you only have the sample. after the induction, you have the conclusion. the object space is the same/duplicated.

For instance, you have a Box labeled "CSI" and solid rectangles in it that are defined "objects with CSI", but what is a blank area in that box supposed to mean? Is it unknown if it has CSI or not, unknown for origin, or both?
you mean objects with ID?. because all objects in that box have CSI, but not all have known [origin].
I've edited in some explanations. ask again if it's not enough

And depending on the above clarification, you have outlined areas (often labeled with "conclusion") that are filled in. Would such a filled in conclusion allow for any unknown properties still?
if the box is filled with "object with ID" it means we induced that all objects in that box probably have ID.

A lot of your box diagrams do not list what the rectangular icons are supposed to represent, so I'm not sure about the point there. I'd rather not guess, so can you please clarify?
edited it in

With respects to the Mt Rushmore example, by what criteria are you able to make any inference (since you said you could)?

Code:
░ = object with natural cause = object with [origin=natural]
▓ = object with ID            = object with [origin=ID]
sample:                               conclusion:
¬CSI            CSI           ¬CSI            CSI
┌─────────────┬─────────────┐  ┌─────────────┬─────────────┐
│░     ░    ▓ │      ▓   ▓  │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓│
│             │   ▓   ▓     │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓│
¬life │     ░  ░ ░  │             │  │░░░░░░░░░░░░▓│▓▓▓▓▓▓▓▓▓▓▓◄─┼─┐
│  ▓      ▓   │     ▓     ▓ │  │░░░░░░░░░░░▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓│ │
│    ▓ ░     ░│ ▓      ▓    │  │░░░░░░░░░░▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓│ │
├─────────────┼─────────────┤  ├─────────────┼─────────────┤ │
│             │             │  │             │             │ │
│             │             │  │             │             │ │
life │             │             │  │   unknown   │   unknown   │ │
│             │             │  │             │             │ │
│             │             │  │             │             │ │
└─────────────┴─────────────┘  └─────────────┴─────────────┘ │
│
Mt. Rushmore is here
That's why we can strongly infer that Mt. rushmore is ID. the underlying sample is full of objects with only ID.

The diagram under your text "after dr. killjoy points out the bias, you have to admit that" is suggesting that any ID inference is not possible. Then how can you infer something like evolution, or are you? (per the diagram, no inference seems possible)
that sample is something ID has put forth to try and prove ID. evolution does not induce itself this way.

evolution, like any other hypothesis, is an induction about observations. newton's law(NL) is already shown to have a bias, so here's a more interesting example based on NL:

Code:
░ = observation does     contradict NL
▓ = observation does not contradict NL
RC = relativistic conditions
sample:                            conclusion:
¬RC              RC                ¬RC              RC
┌───────────────┬───────────────┐  ┌───────────────┬───────────────┐
│▓       ▓      │  ░            │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│░░░░░░░░░░░░░░░│
│  ▓    ▓       │      ░        │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│░░░░░░░░░░░░░░░│
│          ▓    │         ░    ░│  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│░░░░░░░░░░░░░░░│
│        ▓   ▓  │░              │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│░░░░░░░░░░░░░░░│
│        ▓     ▓│          ░    │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│░░░░░░░░░░░░░░░│
│     ▓   ▓     │   ░         ░ │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│░░░░░░░░░░░░░░░│
│           ▓   │       ░       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│░░░░░░░░░░░░░░░│
│  ▓            │ ░        ░    │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│░░░░░░░░░░░░░░░│
└───────────┬───┴──┬────────────┘  └───────────────┴───────────────┘
│      │
object space = all possible experiments/observations/events

before relativistic conditions were measurable and that bias was detected, newtons law was induced to work universally

Lastly, can you please expound or dumb down for me on how you think ID is making a circular argument? Again, the diagram is not understandable to me because the comments and labeling are not clear.
oh thx, I just noticed that it's not a "circular argument", fixed that part.
it's an open quoestion. one that cannot be answered. one that is required to complete the inference, that ID claims to have completed. <- thats all it is.
the circular part comes in - because the open question can be answered from the final conclusion. which means that not answerring the open question = assuming the conclusion is correct. thats why "begging the question" is considered a type of "circular reasoning" fallacies. but since the debater is not aware of it (implicitely circular), it's unfair to call it circular argument.

#### Teax

##### huh?
I hope your browser supports these characters:
Code:
(─) (│) (░) (█) (▓) (►) <-- all brackets have to be filled

[img]http://intpforum.com/attachment.php?attachmentid=2391&stc=1&d=1415616693[/img] <- should look something like this screenshot
-------------------------------------------------------------------------------
diagram is not understandable to me because the comments and labeling are not clear

I tried to combine the 2 graphics.

Code:
░ = object with natural cause = object with [origin=natural]
▓ = object with ID            = object with [origin=ID]

╔═══════════╗
┌─────────────►─╢ induction ╟─┐                 [color=Pink]missing[/color]
│ ╔═══════════╗ ╚═══════════╝ │                 [color=Pink]evidence[/color]
┌────────────┼─╢ induction ║               │                     │
samples:        ▲            ▲ ╚══════════╤╝  conclusions: ▼                     │
¬CSI   │        CSI │            ▼   ¬CSI         │  CSI                ▼
┌───────────┼───┬────────┼──────┐  ┌──┼────────────┬───┼───────────┐     ╔═══╧══════════╗
│┌──────────┴──┐│┌───────┴─────┐│  │┌─┴───────────┐│┌──┴──────────┐│     ║ [life] and   ║
││   ░      ░  │││  ▓   ▓      ││  ││░░░░░░░░░░░░░│││▓▓▓▓▓▓▓▓▓▓▓▓▓││     ║ [origin] are ║
││             │││  ▓   ▓      ││  ││░░░░░░░░░░░░░│││▓▓▓▓▓▓▓▓▓▓▓▓▓││     ║ [url=http://stattrek.com/random-variable/independence.aspx?Tutorial=AP]independent[/url]  ║
¬life ││    ░  ░ ░   │││             ││  ││░░░░░░░░░░░░▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓├┼──┐  ╚═══╤══════════╝
││ ▓      ▓    │││    ▓     ▓  ││  ││░░░░░░░░░░░▓▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓││  │      │
││   ▓ ░     ░ │││▓      ▓     ││  ││░░░░░░░░░░▓▓▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓││  ▼      ▼
│└─────────────┘│└─────────────┘│  │└─────────────┘│└─────────────┘│ ╔╧══════╧═══╗
├───────────────┼───────────────┤  ├───────────────┼───────────────┤ ║ deduction ║
│               │               │  │               │┌─────────────┐│ ╚╤══════════╝
│               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓││  │
│               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓││  ▼
life │               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓├┼──┘
│               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓││
│               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓││
│               │               │  │               │└──┬──────────┘│
└───────────────┴───────────────┘  └───────────────┴───┼───────────┘
this whole side is                 "after induction"   │
"before induction"                                     │
[color=Brown]ID tries to infer this.[/color]

ID tries to infer the property [origin] for objects with life and CSI. It's not possible with induction alone. an additional deduction is required. the necessary independence statement "[life] and [origin] are statistically independent" is required as a premise for the deduction.

No reason is given for "[life] and [origin] are statistically independent". It's the most interesting question: does life has some property that allows it to create CSI in such a way that nonlife could not(=correlation of [life] and [origin])? or is life nothing special from nonlife(=independence of [life] and [origin])? ID, instead, is presenting a sort of misdirection like in a magic trick: nobody even mentions that any of this is required - they just skip this and present their inference in such an obfuscated manner that you don't notice that they're begging the question.

evolution supports a correlation between [life] and [origin]. Let's insert it:

Code:
░ = object with natural cause = object with [origin=natural]
▓ = object with ID            = object with [origin=ID]

╔═══════════╗
┌─────────────►─╢ induction ╟─┐
│ ╔═══════════╗ ╚═══════════╝ │                 [color=SkyBlue]evolution[/color]
┌────────────┼─╢ induction ║               │                     │
samples:        ▲            ▲ ╚══════════╤╝  conclusions: ▼                     │
¬CSI   │        CSI │            ▼   ¬CSI         │  CSI                ▼
┌───────────┼───┬────────┼──────┐  ┌──┼────────────┬───┼───────────┐     ╔═══╧══════════╗
│┌──────────┴──┐│┌───────┴─────┐│  │┌─┴───────────┐│┌──┴──────────┐│     ║ [life] and   ║
││   ░      ░  │││  ▓   ▓      ││  ││░░░░░░░░░░░░░│││▓▓▓▓▓▓▓▓▓▓▓▓▓││     ║ [origin] are ║
││             │││  ▓   ▓      ││  ││░░░░░░░░░░░░░│││▓▓▓▓▓▓▓▓▓▓▓▓▓││     ║ [url=http://stattrek.com/random-variable/independence.aspx?Tutorial=AP][color=SkyBlue]correlated[/color][/url]   ║
¬life ││    ░  ░ ░   │││             ││  ││░░░░░░░░░░░░▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓├┼──┐  ╚═══╤══════════╝
││ ▓      ▓    │││    ▓     ▓  ││  ││░░░░░░░░░░░▓▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓││  │      │
││   ▓ ░     ░ │││▓      ▓     ││  ││░░░░░░░░░░▓▓▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓││  ▼      ▼
│└─────────────┘│└─────────────┘│  │└─────────────┘│└─────────────┘│ ╔╧══════╧═══╗
├───────────────┼───────────────┤  ├───────────────┼───────────────┤ ║ deduction ║
│               │               │  │               │┌─────────────┐│ ╚╤══════════╝
│               │               │  │               ││░░░░░░░░░░░░░││  │
│               │               │  │               ││░░░░░░░░░░░░░││  ▼
life │               │               │  │               ││░░░░░░░░░░░░░├┼──┘
│               │               │  │               ││░░░░░░░░░░░░░││
│               │               │  │               ││░░░░░░░░░░░░░││
│               │               │  │               │└──┬──────────┘│
└───────────────┴───────────────┘  └───────────────┴───┼───────────┘
this whole side is                 "after induction"   │
"before induction"                                     │
cool huh

if you disprove evolution, the best you can do is to say "well then we just don't know".

I hope this clears it all up

#### WookieeB

##### Active Member
Ok, thanks for answering my questions...mostly. There still are a few items you did not answer, but I think I can figure them out now.

my claim: specificity is irrelevant. I heared you bring 2 arguments against evolution. please present these 2 points that I'm only vaguely stating:
• specificity, the fact that a machine is useless without a part. please state, as clear and concise as possible, why this is an argument against evolution. I will then attach my claim as a response. (so I don't attack a strawman)
.
• probability to produce CSI by evolution is too low to be believable. redbaron challenged you here, don't leave him hangin' .

I am planning to get to these, but I want to get through the ID argument structure before we get to specific evidence. But by these statements, you are questioning one of the premises of the ID argument (that non-ID causes cannot produce CSI) and not the argument structure. So, stay tuned...

see the difference?

Yes, I missed a word in there. If I could do it again I would replace ORIGIN with CSI. But at this point it would be irrelevant because I see where I confused the sample with the population. But again, it wont really change the end result...as I will show below.

basic induction:
Code:
▓ = has X
sample = premise                    induction = inference that the space
premise = a couple objects          between the points is same as the points
┌───────────────────────────────┐  ┌───────────────────────────────┐
│▓                      ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓           ▓             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│         ▓                     │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│                      ▓     ▓  │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              ▓                │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│▲    ▓                 ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││    ▲         ▓             ▓ │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▲▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││ ▓  │                         │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└┼────┼─────────────────────────┘  └───────────────┼───────────────┘
│    │                                            │
│    that's an object that we know has X          │
the empty space are objects with unknown X        │
│
the inference tells us that every object probably has X
[/quote]

Got it. OK so far.
What I was doing was confusing and conflating the population size with the sample. I can correct that.
Population = every element in the universe of interest
For the sample size to not be biased, it has to be completely random from among the population.
Sample - a random selection of elements from the population
sampling bias - a sample is collected in such a way that some members of the intended population are less likely to be included than others.

I was making the population all objects where we knew the measurement for CSI, either yesCSI or noCSI. But I realize that was too broad, because for the ID argument we really do not care about noCSI objects, they are not an "element in the universe of interest"

My sample basically was including all of those, but distinquishing groups based on ORIGIN, which I cannot strictly do. I understand that.

But fortunately for me, it ends up not really mattering,..... but I will get to that in a sec.

You though were also making a mistake and focusing on a biased sample too - lifeforms. And by that you are saying ID has a bias, (which I say it doesnt), but you are not identifying the supposed bias correctly.

here's what the ID argument looks like you presented it.
Code:
░ = object with natural cause
▓ = object with ID
sample:                          conclusion:
¬CSI           CSI              ¬CSI           CSI
┌─────────────┬──────────────┐  ┌─────────────┬──────────────┐
│░     ░      │  ▓        ▓  │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ░  ░       │      ▓       │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│    ▓   ░    │              │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│             │   ▓ ▓        │  │░░░░░░░░░░░░▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ▓   ░     ░│           ▓  │  │░░░░░░░░░░░▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│    ░  ░     │              │  │░░░░░░░░░░▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓   ░   │      ▓       │  │░░░░░ partly │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ░          │  ▓    ▓  ▓   │  │░░░░░░░░▓▓▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└─────────────┴──────────────┘  └─────────────┴──────────────┘

This could work, but it is dealing with too diverse of a population. For now, we do not care about elements with noCSI. We only care about yesCSI elements, which would include lifeforms as a subset, or a biased sample.

So, the population (every element in the universe of interest*) is objects with CSI. CSI is measurable.

So in your diagram, we can remove the left of the grouped samples, and be left with -

Code:
░ = object with natural cause
▓ = object with ID
sample:            conclusion:
CSI              CSI
┌──────────────┐  ┌──────────────┐
| ▓        ▓   │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│      ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
|   ▓ ▓        │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│           ▓  │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│      ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ▓    ▓  ▓   │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└──────────────┘  └──────────────┘

The sample is randomly selected elements from the population, no bias. The elements of CSI and ORIGIN are independent. So that would include yesCSI elements that we know the origin, and yesCSI elements that we do not know the origin (which includes lifeforms). But per the premise (which means evidence/data) of the ID argument, all yesCSI elements where we know the origin is always ID, so that is why all the blocks in the sample are "▓"

Now, here is gets sticky.....

after dr. killjoy points out the bias, you have to admit that
Code:
░ = object with natural cause
▓ = object with ID
sample:                               conclusion:
1   ¬CSI      2    CSI         5  ¬CSI       6  CSI
┌─────────────┬─────────────┐  ┌─────────────┬─────────────┐
│░     ░    ▓ │      ▓   ▓  │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓│
│             │   ▓   ▓     │  │░░░░░░░░░░░░░│▓▓▓▓▓▓▓▓▓▓▓▓▓│
¬life │     ░  ░ ░  │             │  │░░░░░░░░░░░░▓│▓▓▓▓▓▓▓▓▓▓▓▓▓│
│  ▓      ▓   │     ▓     ▓ │  │░░░░░░░░░░░▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓│
│    ▓ ░     ░│ ▓      ▓    │  │░░░░░░░░░░▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓│
├─────────────┼─────────────┤  ├─────────────┼─────────────┤
│             │             │  │             │             │
│             │             │  │             │             │
life │             │             │  │   unknown   │   unknown   │
│             │             │  │             │             │
│             │             │  │             │             │
└─────────────┴─────────────┘  └─────────────┴─────────────┘
3              4                7             8

This was your representation of ID. There are two problems here.

1) ID is not distinguishing lifeforms. You have done that but you have not explained why. You are Special pleading - an attempt to cite something (lifeforms) as an exception to a generally accepted rule, principle, etc. without justifying the exception.

ID is making no distiction between yesCSI objects/elements that are life or non-life. That is the rule from the population we are dealing with.

And this is why an inference for MT Rushmore is just like an inference for lifeforms. Per the rules, they should be represented the same in the samples, as they are both yesCSI objects that we do not know the origin (at least before any inference is made).

2) Even if we allow for your special case (lifeforms) and we are looking at noCSI and yesCSI objects at the same time, you represented the boxes wrong. Box #3 and Box #7 would never have any icon in them, because lifeforms = yesCSI per the rules, there is no such thing as life = noCSI. So your second vertical level is fallacious.

What your vertical distinction should be is between known ORIGIN and unknown ORIGIN

So to correct your boxes, ID should be represented as -
Code:
░ = object with natural cause
▓ = object with ID
sample:           conclusion:
1    CSI         3    CSI
┌─────────────┐  ┌─────────────┐
│▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
│▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
known     │▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
origin    │▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
│▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
├─────────────┤  ├─────────────┤
│             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
unknown   │             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
origin    │             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│<-- the inference is to 'what is the origin'
(including │             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│    so it would not stay unknown
lifeforms)│             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
└─────────────┘  └─────────────┘
2                4

Now, when we come upon something with yesCSI (like Mt Rushmore) where we do not initially know the origin, because it has yesCSI we infer ID. When we finally discover the factual origin (like in man-made Mt Rushmore), we confirm the inference. Whether Mt Rushmore is alive or not is irrelevant. What is relevent is whether it has yesCSI or noCSI.

When an archaeologist comes upon some new artifact, they do not know the origin initially. Take something simple like a rock that initially appears (before any detailed study) in the form of an arrowhead. They can then study it to define in detail properties (evidence) about the object. Do the properties indicate the shape matches a specification, yesCSI (an arrowhead has specific requirements: like the rock chipped to a point, chipped edges that are sharp, edges form is flared out in a relatively symmetrical manner, etc,...). Or are the properties of the object lacking a specification, noCSI, and more likely due to random natural forces. Archaelogy uses the ID inference - if yesCSI then it is an arrowhead, ID. If noCSI, then it is just a rock, not ID.
(Please note, that an artifact doesnt have to specify to an arrowhead. There are other specifications that would still infer ID, like maybe the rock was shaped as a hook, or smooth and rounded for crushing food, or..... But there would be enough of a specification to the object to distinquish it as an ID object vs random nature, and enough to distinguish it among different ID objects (arrowhead vs hook))

Forensic science also uses an ID inference. They know an event occurred, but do not know the origin or specifics initially. As they study the evidence, they can rule on specified vs random causes and then infer ID vs nature. But the evidence around CSI is what supports infer:specification or infer:random, not some special property like life vs not-life.

When it comes to the ID inference and lifeforms, it works the same way. Does the item have yesCSI or noCSI. If yesCSI, in all cases where we have come across a known origin, it has been ID, thus we can infer ID as the origin.

So, while I was writing the above, you wrote.....

Code:
░ = object with natural cause = object with [origin=natural]
▓ = object with ID            = object with [origin=ID]

╔═══════════╗
┌─────────────►─╢ induction ╟─┐                 [color=Pink]missing[/color]
│ ╔═══════════╗ ╚═══════════╝ │                 [color=Pink]evidence[/color]
┌────────────┼─╢ induction ║               │                     │
samples:        ▲            ▲ ╚══════════╤╝  conclusions: ▼                     │
¬CSI   │        CSI │            ▼   ¬CSI         │  CSI                ▼
┌───────────┼───┬────────┼──────┐  ┌──┼────────────┬───┼───────────┐     ╔═══╧══════════╗
│┌──────────┴──┐│┌───────┴─────┐│  │┌─┴───────────┐│┌──┴──────────┐│     ║ [life] and   ║
││   ░      ░  │││  ▓   ▓      ││  ││░░░░░░░░░░░░░│││▓▓▓▓▓▓▓▓▓▓▓▓▓││     ║ [origin] are ║
││             │││  ▓   ▓      ││  ││░░░░░░░░░░░░░│││▓▓▓▓▓▓▓▓▓▓▓▓▓││     ║ [url=http://stattrek.com/random-variable/independence.aspx?Tutorial=AP]independent[/url]  ║
¬life ││    ░  ░ ░   │││             ││  ││░░░░░░░░░░░░▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓├┼──┐  ╚═══╤══════════╝
││ ▓      ▓    │││    ▓     ▓  ││  ││░░░░░░░░░░░▓▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓││  │      │
││   ▓ ░     ░ │││▓      ▓     ││  ││░░░░░░░░░░▓▓▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓││  ▼      ▼
│└─────────────┘│└─────────────┘│  │└─────────────┘│└─────────────┘│ ╔╧══════╧═══╗
├───────────────┼───────────────┤  ├───────────────┼───────────────┤ ║ deduction ║
│               │               │  │               │┌─────────────┐│ ╚╤══════════╝
│               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓││  │
│               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓││  ▼
life │               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓├┼──┘
│               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓││
│               │               │  │               ││▓▓▓▓▓▓▓▓▓▓▓▓▓││
│               │               │  │               │└──┬──────────┘│
└───────────────┴───────────────┘  └───────────────┴───┼───────────┘
this whole side is                 "after induction"   │
"before induction"                                     │
[color=Brown]ID tries to infer this.[/color]

ID tries to infer the property [origin] for objects with life and CSI.

No, it does not. I have been saying this from the beginning.

ID tries to infer the property [origin] for objects with CSI. Objects with CSI include life. That was a premise from the beginning.

So, in your new diagram, anything element/object the boxes directly above the words "this whole side is" and "after induction" are illogical.

It's not possible with induction alone. an additional deduction is required. the necessary independence statement "[life] and [origin] are statistically independent" is required as a premise for the deduction.

Okay.,,,,,? Yes, I suppose so.

No reason is given for "[life] and [origin] are statistically independent". It's the most interesting question: does life has some property that allows it to create CSI in such a way that nonlife could not(=correlation of [life] and [origin])? or is life nothing special from nonlife(=independence of [life] and [origin])? ID, instead, is presenting a sort of misdirection like in a magic trick: nobody even mentions that any of this is required - they just skip this and present their inference in such an obfuscated manner that you don't notice that they're begging the question.

Again, life is not the particular property ID is talking about. It is the special case YOU are trying to push.

ID is talking about objects with CSI, which evidentially we have established includes lifeforms, but it also includes non-life objects like Mt Rushmore, or a car, or a new pyramid found in Egypt, or bullet-holes in a body. There is no misdirection, except by you trying to make a special case out of life.

evolution supports a correlation between [life] and [origin]
Excuse me?!? How so?

if you disprove evolution, the best you can do is to say "well then we just don't know".

NO, if you disprove evolution, the inference stands as it is, cause the inference is already operating under the assumption that evolution is insufficient to produce objects with CSI.

#### Teax

##### huh?
What your vertical distinction should be is between known ORIGIN and unknown ORIGIN
note that these 2 graphs try to represent exactly the same thing, but the right one contradicts itself, and is redundant, so use the left representation.

in the right graph, the box4, which is the same box as box2, is simultaneously labelled "unknown origin" and contains '▓' defined to have [origin=yesID] -> contradiction.

Code:
▓ = object with ID            = [origin=yesID]
space = [origin=unknown]

sample:            conclusion:                    sample:           conclusion:
CSI              CSI                        1    CSI         3    CSI
┌──────────────┐  ┌──────────────┐               ┌─────────────┐  ┌─────────────┐
| ▓        ▓   │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
│      ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│     known     │▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
|   ▓ ▓        │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│     origin    │▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
│           ▓  │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │▓▓▓▓▓▓▓▓▓▓▓▓▓│  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               ├─────────────┤  ├─────────────┤
│ ▲    ▓       │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
│ │▓    ▓  ▓   │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓│     unknown   │             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
└─┼────────────┘  └──────────────┘     origin    │             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│<-- the inference is to 'what is the origin'
│                                   (including │             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│    so it would not stay unknown
the empty space are                   lifeforms)│             │  │▓▓▓▓▓▓▓▓▓▓▓▓▓│
objects with unknown X                           └─────────────┘  └─────────────┘
space = [origin=unknown]                         2                4

Excuse me?!? How so?
applied to CSI, evolution is a theory that - states that the mathematical model of small changes and statistical selection accounts for CSI in life (yesLife-noID). non-living objects do not follow that model and require an outside source of CSI (noLife-yesID). consequentially evolution supports a correlation between [origin] and [life] for CSI objects. trivial right?

Got it. OK so far.
What I was doing was confusing and conflating the population size with the sample. I can correct that.
Population = every element in the universe of interest
For the sample size to not be biased, it has to be completely random from among the population.
Sample - a random selection of elements from the population
sampling bias - a sample is collected in such a way that some members of the intended population are less likely to be included than others.
the "size" is a typo? the rest is correct, yes, if you interpreted it correct. <- you didn't. you make the same mistake with samples below again.

But fortunately for me, it ends up not really mattering,.....
no surprise =p
The sample is randomly selected elements from the population, no bias. The elements of CSI and ORIGIN are independent. So that would include yesCSI elements that we know the origin, and yesCSI elements that we do not know the origin (which includes lifeforms).
The sample is randomly selected apples from the basket, no bias. -------beeeeeeeeeeeep-------- So that would include apples that we tasted, and apples that we didn't taste (which includes green apples).
X = [origin]
X = sweetness of apples
the objects with unknown X are the ones you are making the conclusion about. thus a sample by definition cannot contain objects with unknown X. otherwise it's an explicit circular argument.

how did you come to the conclusion that its random? show me that no memebers are less likely to be included than others. the objects you are allowed to pick your sample from are all non-life objects.

I have no idea what you mean by independent-elements.

Code:
▓ = sweet
sample:                             population:
red + apple      green + apple     red + apple     green + apple
┌───────────────┬───────────────┐  ┌───────────────┬───────────────┐
│▓       ▓      │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│  ▓    ▓       │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│          ▓    │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│    stays      │
│        ▓   ▓  │               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│   unknown     │
│        ▓     ▓│               │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│        ▲  ▓   │          ▲    │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
│  ▓     │      │          │    │  │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│               │
└────────┼──────┴──────────┼────┘  └───────────────┴───────────────┘
│                 │
these ▓ are your sample   │
you tasted them           │
[color=red]none of this can be in your sample[/color]
do you agree? if not, deduce the contradiction and present a "good" analogy that illustrates that my red statements are wrong/lead to ridiculous consequences.

ID tries to infer the property [origin] for objects with CSI. Objects with CSI include life. That was a premise from the beginning.
once you realize that, you should be able to see that ID indeed like you said tries to induce all CSI. It's not allowed to.

not considering this bias, in lamens terms means: trying to infer [origin] for all of the population at the same time = in 1 induction step, without giving any reason to believe that all of the population actually has the same [origin].

you are getting this basic thing wrong, and the rest you said is based on this faulty premise. it's like an avalanche.
http://www.youtube.com/watch?v=BpNxRsIoN58

#### WookieeB

##### Active Member
*sigh* Lets see if I can make this as simple as I can, using YOUR diagrams.

But I have to edit a little. I have to add another property "Y" to minimally match your other diagrams.

Code:
  Population - object has Y
▓ = has X
sample = premise                         induction = inference that the space
premise = a couple objects               between the points is same as the points
Y                                        Y
┌───────────────────────────────┐       ┌───────────────────────────────┐
│▓                      ▓       │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓           ▓             │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│         ▓                     │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│                      ▓     ▓  │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              ▓                │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│▲    ▓                 ▓       │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││    ▲         ▓             ▓ │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▲▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││ ▓  │                         │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└┼────┼─────────────────────────┘       └───────────────┼───────────────┘
│    │                                                 │
│    that's an object that we know has X (but has Y)   │
the empty space are objects with unknown X (but has Y) │
│
the inference tells us that every object probably has X
Is the above a valid induction?

To demonstrate other properties, you gave this diagram...
Code:
░ = object with natural cause = object with [origin=natural]
▓ = object with ID            = object with [origin=ID]

╔═══════════╗
┌─────────────►─╢ induction ╟─┐
│ ╔═══════════╗ ╚═══════════╝ │                 [color=SkyBlue]evolution[/color]
┌────────────┼─╢ induction ║               │                     │
samples:        ▲            ▲ ╚══════════╤╝  conclusions: ▼                     │
¬CSI   │        CSI │            ▼   ¬CSI         │  CSI                ▼
┌───────────┼───┬────────┼──────┐  ┌──┼────────────┬───┼───────────┐     ╔═══╧══════════╗
│┌──────────┴──┐│┌───────┴─────┐│  │┌─┴───────────┐│┌──┴──────────┐│     ║ [life] and   ║
││   ░      ░  │││  ▓   ▓      ││  ││░░░░░░░░░░░░░│││▓▓▓▓▓▓▓▓▓▓▓▓▓││     ║ [origin] are ║
││             │││  ▓   ▓      ││  ││░░░░░░░░░░░░░│││▓▓▓▓▓▓▓▓▓▓▓▓▓││     ║ [url=http://stattrek.com/random-variable/independence.aspx?Tutorial=AP][color=SkyBlue]correlated[/color][/url]   ║
¬life ││    ░  ░ ░   │││             ││  ││░░░░░░░░░░░░▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓├┼──┐  ╚═══╤══════════╝
││ ▓      ▓    │││    ▓     ▓  ││  ││░░░░░░░░░░░▓▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓││  │      │
││   ▓ ░     ░ │││▓      ▓     ││  ││░░░░░░░░░░▓▓▓│││▓▓▓▓▓▓▓▓▓▓▓▓▓││  ▼      ▼
│└─────────────┘│└─────────────┘│  │└─────────────┘│└─────────────┘│ ╔╧══════╧═══╗
├───────────────┼───────────────┤  ├───────────────┼───────────────┤ ║ deduction ║
│               │               │  │               │┌─────────────┐│ ╚╤══════════╝
│               │               │  │               ││░░░░░░░░░░░░░││  │
│               │               │  │               ││░░░░░░░░░░░░░││  ▼
life │               │               │  │               ││░░░░░░░░░░░░░├┼──┘
│               │               │  │               ││░░░░░░░░░░░░░││
│               │               │  │               ││░░░░░░░░░░░░░││
│               │               │  │               │└──┬──────────┘│
└───────────────┴───────────────┘  └───────────────┴───┼───────────┘
this whole side is                 "after induction"   │
"before induction"                                     │
cool huh

I dont feel like editing this picture, but I'm focusing just on the top horizontal line of boxes, with the "induction" lines. To you, these are valid inductions, right?

#### Teax

##### huh?
But I have to edit a little.
sure edit as much as you want! pictures are more fun than just text.

I have to add another property "Y" to minimally match your other diagrams.

yep. I'm gonna clear up a couple of things just to make it more to the point:

Code:
  Population - object has Y
▓ = has X
sample = premise                         induction = inference that the space
premise = a couple objects               between the [color=SkyBlue]sample elements[/color] is same as the
[color=SkyBlue]with known X[/color]                       [color=SkyBlue]sample elements[/color]

Y                                        Y
┌───────────────────────────────┐       ┌───────────────────────────────┐
│▓                      ▓       │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓           ▓             │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│         ▓                     │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│                      ▓     ▓  │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              ▓                │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│▲    ▓                 ▓       │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││    ▲         ▓             ▓ │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▲▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││ ▓  │                         │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└┼────┼─────────────────────────┘       └───────────────┼───────────────┘
│    │                                                 │
│    that's an object that we know has X ([color=SkyBlue]and[/color] has Y)   │
the empty space are objects with unknown X ([color=SkyBlue]and[/color] has Y) │
│
the inference tells us that every object probably has X

Is the above a valid induction?
sure, the induction will be valid. (but like with any induction, only untill someone points out a bias or counterexample)

To demonstrate other properties, you gave this diagram... I dont feel like editing this picture, but I'm focusing just on the top horizontal line of boxes, with the "induction" lines. To you, these are valid inductions, right?
yep. both those inductions in the upper line are analogous to your diagram, with variables mapped to specific attributes. here's the mapping:

the upper left induction is:
X = ¬ID
Y = ¬life + ¬CSI

the upper right induction is:
X = ID
Y = ¬life + CSI

so far so good, proceed

#### WookieeB

##### Active Member
Ok, better, but there is one more item that needs clarification.

Code:
  Population - object has Y
▓ = has X
sample = premise                         induction = inference that the space
premise = a couple objects               between the [color=SkyBlue]sample elements[/color] is same as the
[color=SkyBlue]with known X[/color]                       [color=SkyBlue]sample elements[/color]

Y                                        Y
┌───────────────────────────────┐       ┌───────────────────────────────┐
│▓                      ▓       │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓           ▓             │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│         ▓                     │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│                      ▓     ▓  │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              ▓                │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│▲    ▓                 ▓       │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││    ▲         ▓             ▓ │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▲▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││ ▓  │                         │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└┼────┼─────────────────────────┘       └───────────────┼───────────────┘
│    │                                                 │
│    that's an object that we know has X ([color=SkyBlue]and[/color] has Y)   │
the empty space are objects with unknown X ([color=SkyBlue]and[/color] has Y) │
│
the inference tells us that every object probably has X

I just want to be as clear as possible. In in the left box, a sample element is an object represented by an individual colored rectangle icon, = has X (and has Y). Collectively these icons are the sample elements.

In in the left box, an empty space is also an object, but not a sample element. It is an object that may or may not have X (but regardless if it has X OR has NOT X, it still has Y). X is for the moment unknown. I just want to make it clear that though it is represented by an 'empty space' it does not mean there is no object, but just that the object property for X is unknown. It also does not represent an object that is = NOT X, as that object property would be represented by a different (colored rectangle) icon. So the left box is FULL of objects.

For the right box, you wrote: "induction = inference that the [empty] space between the sample elements is same as the sample elements"

I added "[empty]" to clarify and match it up to your wording on the left box descriptions. Perhaps we can rename those objects to "unknown-x-spaces" or "?x?-spaces" or similar. I'm just trying to alleviate any impression that one might have looking at the picture and think the "space" means there is nothing there, or that there is no X there. (I don't want to necessairly substitute a different icon in for the empty space icon)

The induction is that for any unknown-X objects, we can infer they probably have X because all the other non-unknown-X objects all have X.

So perhaps we could rewrite - "induction = inference that the unknwon-x spaces between the sample elements spaces are the same as the sample elements"

Correct?

#### Teax

##### huh?
In in the left box, a sample element is an object represented by an individual colored rectangle icon, = has X (and has Y). Collectively these icons are the sample elements.
affirmative.

In in the left box, an empty space is also an object, but not a sample element. It is an object that may or may not have X (but for either it still has Y). X is for the moment unknown. I just want to make it clear that though it is represented by an 'empty space' it does not mean there is no object, but just that the object property for X is unknown. It also does not represent an object that is = NOT X, as that object property would be represented by a different (colored rectangle) icon. So the left box is FULL of objects.
let's assume that yes - because it does not matter for the inferences. and makes it easier to think about this.

the conclusion sais: all objects probably have X. which is not a contradiction, even if there are no objects to begin with haha.

technically: space = "area where an object might reside"

For the right box, you wrote: "induction = inference that the [empty] space between the sample elements is same as the sample elements"

I added "[empty]" to clarify and match it up to your wording on the left box descriptions. Perhaps we can rename those objects to "unknown-x-spaces" or "?x?-spaces" or similar. I'm just trying to alleviate any impression that one might have looking at the picture and think the "space" means there is nothing there, or that there is no X there. (I don't want to necessairly substitute a different icon in for the empty space icon)
"no X there" = ¬X

right the "unknown-x-space" definetly does not mean "object has ¬X". it rather means "unknown whether it's ¬X or X". it could have ¬X... but we don't know.

The induction is that for any unknown-X objects, we can infer they probably have X because all the other non-unknown-X objects all have X.
yes, but incomplete.

since you wrote a "because", we need 2 reasons to formally induce that. the second is:
• we assume the non-unknown-X objects are picked randomly
= for every other variable Z, we have objects that have Z and ¬Z, as part of the sample = with known X.

the randomness of the ▓ in that box is intended to visuallize the "random picking". A biased sample could be visualized e.g. like this:

Code:
┌───────────────────────────────┐
│  ▓ ▓                          │
│▓    ▓                         │
│  ▓   ▓                        │
│▓   ▓                          │
│ ▓  ▓                          │
└───────────────────────────────┘

So perhaps we could rewrite - "induction = inference that the unknwon-x spaces between the sample elements spaces are the same as the sample elements"
spaces typo? otherwise yeah, it sounds even more to the point.

Code:
  Population - object has Y
"▓" = has X
" " = unknown, could be X or ¬X

sample = premise                         induction = inference that the unknown-x spaces
premise = a couple randomly picked       between the sample elements are same as the
objects with known X                  sample elements
(and every possible new Z)

Y                                        Y
┌───────────────────────────────┐       ┌───────────────────────────────┐
│▓                      ▓       │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│     ▓           ▓             │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│         ▓                     │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│                      ▓     ▓  │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│              ▓                │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
│▲    ▓                 ▓       │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││    ▲         ▓             ▓ │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▲▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
││ ▓  │                         │       │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓│
└┼────┼─────────────────────────┘       └───────────────┼───────────────┘
│    │                                                 │
│    that's an object that we know has X (and has Y)   │
the empty space are objects with unknown X (and has Y) │
│
the inference tells us that every object probably has X

hope it helps

#### Cherry Cola

##### Banned
at some point a black hole is going to be created as a result of this discussion and then we will all die