Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Final Part: Modern ID Theory Demonstrates the Limits of Naturalistic Process; Jillery's Defense is an Argument from Ignorance

49 views
Skip to first unread message

T Pagano

unread,
May 20, 2018, 3:25:03 PM5/20/18
to talk-o...@moderators.isc.org
On Mon, 14 May 2018 15:49:58 -0400, jillery wrote:

> On Sun, 13 May 2018 16:24:41 GMT, T Pagano <notmya...@dot.com> wrote:
>
>>On Fri, 11 May 2018 13:08:44 -0400, jillery wrote:
>>
>>> On Fri, 11 May 2018 12:31:17 GMT, T Pagano <notmya...@dot.com>
>>> wrote:
>>>
>>>>On Thu, 10 May 2018 19:14:02 -0500, Freon96 wrote:

snipped and previously covered.

>
>>8. Jillery argued that "Even if material cause is insufficient in the
>>present, that does not imply it will be insufficient in the future."
>>Unfortunately such a claim makes the limits of material processes
>>untestable and hence the claim is unscientific.
>
>
> Ok, you caught Jillery in a poorly phrased comment. Jillery meant to
> say that even if a present explanation of material cause was
> insufficient, that doesn't imply a sufficient material explanation won't
> be found in the future. Nor does it make material cause unprovable. It
> simply notes the limits of what is known at any one time.

This is substantially identical to the quote of yours I criticized above.

1. In both instances you argue that the existence of some unknown
naturalistic explanation may always be discovered in the future. Being
out of our reach such claims are untestable and therefore unscientific.

2. Both instances also suffers from the classic, fallacious "argument
from ignorance." Your argument's appeal to ignorance of our current
understanding as the lone premise in an argument justifying the existence
of unknown naturalistic explanations. This makes your argument invalid.

3. Finally Dembski's Theory (when justified) rules out the entire class
of naturalistic explanations not just the ones we know about. It does so
with a rigorous application of both probability and complexity theory.
Both well accepted and settled sciences.

4. For example, if a SETI researcher were to read off the first 500
prime numbers in order off of some incoming signal from Alpha Centuri,
Dembski's theory would characterize the specification and complexity of
this information such that it would be unreachable by any logically
possible naturalistic explanation.


> For example, preformationism was once asserted as a material cause of
> heredity. Further experiments proved it to be incorrect, but that
> doesn't mean there's no material cause for heredity.

This example proves that theories are provisional and are sometimes
superseded in the future. It does not remedy your fallacious argument
from ignorance made above.






>

Burkhard

unread,
May 20, 2018, 6:05:02 PM5/20/18
to talk-o...@moderators.isc.org
don't you find it a trifle odd that you have problems with theories that
have unrestricted universal quantifiers about such a mundane thing as
physical objects ("all swans are white") which according to your reading
of Hume can't possibly be justified, and now you extol a "theory" that
quantifies not only over all past, present and future objects, but also
all past, present and future theories?

And no, there is no rigorous application of probability or complexity
theory anywhere in Behe, just enough hand waving to power an entire
windfarm.

Burkhard

unread,
May 20, 2018, 6:10:02 PM5/20/18
to talk-o...@moderators.isc.org
sorry, meant Dembski of course, Did we have a laugh in the forensic
statistics reading group.

Mark Isaak

unread,
May 20, 2018, 11:45:02 PM5/20/18
to talk-o...@moderators.isc.org
On 5/20/18 12:23 PM, T Pagano wrote:
> [...]
> 4. For example, if a SETI researcher were to read off the first 500
> prime numbers in order off of some incoming signal from Alpha Centuri,
> Dembski's theory would characterize the specification and complexity of
> this information such that it would be unreachable by any logically
> possible naturalistic explanation.

And Dembski, like you, would be wrong. A justifiable conclusion would
be that the information would be unreachable by any *known* natural
process. The conclusion you suggest is just plain loopy.

--
Mark Isaak eciton (at) curioustaxonomy (dot) net
"I think if we ever reach the point where we think we thoroughly
understand who we are and where we come from, we will have failed."
- Carl Sagan

jillery

unread,
May 22, 2018, 3:05:02 AM5/22/18
to talk-o...@moderators.isc.org
On Sun, 20 May 2018 19:23:43 GMT, T Pagano <notmya...@dot.com>
wrote:
>>
>>>8. Jillery argued that "Even if material cause is insufficient in the
>>>present, that does not imply it will be insufficient in the future."
>>>Unfortunately such a claim makes the limits of material processes
>>>untestable and hence the claim is unscientific.
>>
>>
>> Ok, you caught Jillery in a poorly phrased comment. Jillery meant to
>> say that even if a present explanation of material cause was
>> insufficient, that doesn't imply a sufficient material explanation won't
>> be found in the future. Nor does it make material cause unprovable. It
>> simply notes the limits of what is known at any one time.
>
>This is substantially identical to the quote of yours I criticized above.


Once again, you removed the context of my comment, which makes it
easier to change your criticism. That's a stupid and dishonest thing
to do.


>1. In both instances you argue that the existence of some unknown
>naturalistic explanation may always be discovered in the future. Being
>out of our reach such claims are untestable and therefore unscientific.


Incorrect. My revised comments addresses your original criticism,
which you conveniently deleted. My original comment refers to the
same cause. My revised comment says a different cause may be found.
There's a difference.

And how do you test "unknown explanations" anyway? Do you even think
about what you type before you post it?


>2. Both instances also suffers from the classic, fallacious "argument
>from ignorance." Your argument's appeal to ignorance of our current
>understanding as the lone premise in an argument justifying the existence
>of unknown naturalistic explanations. This makes your argument invalid.


Your comment above is based on a unique and incoherent meaning of
"argument from ignorance". It does not mean what you think it means:

From Wikipedia:
****************************
It asserts that a proposition is either true or false because of lack
or absence of evidence or proof to the contrary.
****************************

I make no claim here that any theory is correct or false. You are the
one who claims that ID is true by default. I point out that's a false
dichotomy.


>3. Finally Dembski's Theory (when justified) rules out the entire class
>of naturalistic explanations not just the ones we know about. It does so
>with a rigorous application of both probability and complexity theory.
>Both well accepted and settled sciences.
>
>4. For example, if a SETI researcher were to read off the first 500
>prime numbers in order off of some incoming signal from Alpha Centuri,
>Dembski's theory would characterize the specification and complexity of
>this information such that it would be unreachable by any logically
>possible naturalistic explanation.
>
>
>> For example, preformationism was once asserted as a material cause of
>> heredity. Further experiments proved it to be incorrect, but that
>> doesn't mean there's no material cause for heredity.
>
>This example proves that theories are provisional and are sometimes
>superseded in the future. It does not remedy your fallacious argument
>from ignorance made above.


Of course, I posted no fallacious argument, from ignorance or
otherwise.

Of course theories are provisional, As I said, new information is
always being discovered, which may require existing theories to be
modified, or even overturned. That's why nobody can say they have
accounted for all cases, as a matter of principle. Which means even
you can't say you have eliminated all possible material cause, and so
you can't assume ID by default, as a matter of principle.

--
I disapprove of what you say, but I will defend to the death your right to say it.

Evelyn Beatrice Hall
Attributed to Voltaire

T Pagano

unread,
May 24, 2018, 9:30:03 AM5/24/18
to talk-o...@moderators.isc.org
Firstly this doesn't rescue Jillery's argument-from-ignorance.

The triviality of the content of, "all swans are white," is irrelevant to
Hume's problem of induction. Instead Hume addressed both the ability to
justify its truth value. Hume reported that replacing truth with
probably truth didn't remediate the problem.

Dembski/Behe didn't make any existential claims, but theorized limits of
a cause-effect relationship. Similarly laws of nature tells us the
limits of cause-effect relationships not what will/will not exist.

>
> And no, there is no rigorous application of probability or complexity
> theory anywhere in Behe, just enough hand waving to power an entire
> windfarm.

I never claimed that Behe made a mathematically rigorous argument. His
is a theory routed in the limits of observable, real-world, biological
processes. No evolutionist has produced evidence that any known
biological process can explain the origin of the bacterial flagellum.
Jillery's solution is to claim that our ignorance of the existence of a
biological process refutes Behe. As you well know "ignorance" is not a
valid solution to anything.

Dembski generalized the results of Behe using a fairly novel and rigorous
argument about the limits of material causation, in general, (using
probability and complexity theory). Dembski's, "The Design Inference,"
was peer-reviewed and published by Cambridge University Press in 1998.
Elsberry and Wilkins spent 50 pages critiquing Dembski's theory without
any fatal flaws. Their argument effectively reduced to Jillery's
argument from ignorance. Care to improve on their argument by actually
reading Dembski's work rather than stabbing the usual strawman?


snipped.

Earle Jones

unread,
May 24, 2018, 1:20:03 PM5/24/18
to talk-o...@moderators.isc.org
*
T Pagano:

Do you agree with Dembski, when he said this:

"Christ is indispensable to any scientific theory, even if
its practitioners do not have a clue about him."

"My thesis is that all disciplines find their completion in Christ
and cannot be properly understood apart from Christ."

--William Dembski, 'Intelligent Design', p 206

earle
*


T Pagano

unread,
May 24, 2018, 3:30:02 PM5/24/18
to talk-o...@moderators.isc.org
Since the quote has no context it's not clear exactly what Dembski
means. Regardless "Christ" and "theological doctrines" don't appear
anywhere in Dembski's two principle works (The Design Inference; No Free
Lunch) explaining this theory. So his quotes do NOT mean that his theory
requires presuming God or any theological statements.

Furthermore how are Dembski's quotes qualitatively different from Richard
Dawkins, writing in the introduction to his "Blind Watchmaker" that
neoDarwinian theory made him a self fulfilled atheist? In both instances
Dembski and Dawkins are making comments "about" theories in the light of
their world views. In neither case were their world views integral parts
of the theories they support.

jillery

unread,
May 24, 2018, 11:00:03 PM5/24/18
to talk-o...@moderators.isc.org
On Thu, 24 May 2018 13:29:44 GMT, T Pagano <notmya...@dot.com>
wrote:


>Firstly this doesn't rescue Jillery's argument-from-ignorance.


Firstly, Jillery doesn't need rescuing from your imaginary molehills,
especially when you ran away from backing up your claims.

Burkhard

unread,
May 25, 2018, 5:20:03 AM5/25/18
to talk-o...@moderators.isc.org
Jillery did not make an argument from ignorance she accused the IDlers
of building their theories on one

>
> The triviality of the content of, "all swans are white," is irrelevant to
> Hume's problem of induction. Instead Hume addressed both the ability to
> justify its truth value. Hume reported that replacing truth with
> probably truth didn't remediate the problem.

And you totally misunderstand the point I was making. If you are
concerned about an inability to justify even such a relatively simple
claim, then you should be even more concerned about the much more far
reaching universal quantifiers in Dembski's theory that ranges not just
over objects, but over all possible theories.

As an aside your reading of Hume on probability is ahistorical, at the
time he was writing it meant something rather different,


>
> Dembski/Behe didn't make any existential claims, but theorized limits of
> a cause-effect relationship. Similarly laws of nature tells us the
> limits of cause-effect relationships not what will/will not exist.

No idea what that's supposed to mean, or why it matters, but both make
universal statement (or rather negative existential statement), which
would of course be prime examples for Humean induction scepticis of the
type you argue

>
>>
>> And no, there is no rigorous application of probability or complexity
>> theory anywhere in Behe, just enough hand waving to power an entire
>> windfarm.
>
> I never claimed that Behe made a mathematically rigorous argument.

I corrected myself in my dorect follow u - it was Dembski I meant

His
> is a theory routed in the limits of observable, real-world, biological
> processes. No evolutionist has produced evidence that any known
> biological process can explain the origin of the bacterial flagellum.
> Jillery's solution is to claim that our ignorance of the existence of a
> biological process refutes Behe. As you well know "ignorance" is not a
> valid solution to anything.
>
> Dembski generalized the results of Behe using a fairly novel and rigorous
> argument about the limits of material causation, in general, (using
> probability and complexity theory).

He made up some numbers that look to people with no knowledge of
probability theory bit like probabilistic agents, and he uses the word
"complex", but not in the way of any complexity theory

Dembski's, "The Design Inference,"
> was peer-reviewed and published by Cambridge University Press in 1998.

As was pointed out to you before, books are to peer reviewed in that
sense the reviewer only sees the chapter outline and a one page
description, and is then asked if they see a market for such a book, if
it is better marketed at philosophers or theologians, if judging from
the table of content there is an obvious omission, or something that
could be cut, and if there are similar books out there that the book
would compete with.

They don;t read or judge the content. That happens in the post
publication peer review in book reviews in journals, which have been
pretty universally scathing about this abuse of math.


> Elsberry and Wilkins spent 50 pages critiquing Dembski's theory without
> any fatal flaws. Their argument effectively reduced to Jillery's
> argument from ignorance. Care to improve on their argument by actually
> reading Dembski's work rather than stabbing the usual strawman?

I not only read it, I gave it to our research group in forensic
probability, having bee lured b the description that it offers a new
method for forensics reasoning - I did not know Dembski at the time,
and did not even know that some people don't believe in the ToE. That's
what brought me to Ta Origins, as a matter of fact. None of us saw
anything of value in the book, it is not how people in any established
discipline work, and for very good reasons.

>
>
> snipped.
>

0 new messages