Irreducible Functionality and Complexity

22 views
Skip to first unread message

backspace

unread,
Jan 27, 2011, 3:03:12 PM1/27/11
to
http://scratchpad.wikia.com/wiki/Irreducible_Functionality

=== Irreducible optimized functionality ===
The achieve the function or purpose of catching mouse needs five
parts. There can't be less than five parts or the ''optimum''
functionality won't be achieved.

Removing two parts as Kenneth Miller does achieves the fuctionality of
a powerful paper clip: it no longer can catch mouse automatically.
Miller is equivocating between different types of functionalities with
the word "complexity". A better term for Behe's (Irreducbile
Complexity)IC is '''Irreducible optimized functionality'''. The
purpose of the mouse trap is to catch mouse ''unattended'' which as
far as mouse trap engineers can determine needs a minimum of five
parts. Removing the mouse base and stapling it to the floor doesn't
remove the ''purpose'' of the original base but replaces one type of
base for another and it removes the mobility of the mouse trap thus
reducing its overall purposefulness: Reduced functionality.

Same with the flagellum, it needs 40 parts to swim through the fluid:
to achieve the purpose of ''Irreducbile Optimized
Functionality(IOF)''. The type III secretory mechanism which injects
toxins needs x parts to achieve a different IOF. purpose. They both
use some of the same shapes, they both need a minimum of 40 and x
parts respectively in order to '''optimize''' their functionality.
Some IC systems could still function with less parts but like the
mouse trap stapled to the ground at the cost of reduced functionality.

With less than 40 parts , a Calvin and Hobb's creation is made, more a
Rube Goldberg contraption. Both the toxin and flagellum use a round
shape reducing friction the most. A watch and turbine use round shapes
because it achives optimized fuctionality.

== Austetics ==
http://en.wikipedia.org/wiki/Aesthetics. Human designers making
watches don't always strive for the most optimized solution with
minimal parts but a combination of Rube Goldberg(RG) and
optimality(more parts than needed).
Extending the same notion to IOF three notions are proposed:
* 1) There will exist biological machines that are purely
IOF(irreducibly optimized functional).
* 2) There will exist biological machines that combine IF(irreducible
functionality) with Rube Goldberg geniality to make a statement of
austetics: ''Irreducible functionality Rube Goldberg''.
* 3) The larger the animal the more we tend to have IFRG, as we go
smaller it will become more IOF until we reach the smallest micro
machines such as as bacterial flagellum which will be fully IOF and
not IFRG.

Darwin in one of his passages on the http://en.wikipedia.org/wiki/Tufted_Titmouse
speculated that larger feet would have been a better design for the
bird. He insinuated that if it was made by God ,that he didn't know
what he was doing. Instead we should view every single biological
design as oscillating between IOF and IFRG. The titmouse is optimzed
for both its functions of grasping twigs and flying: feet that are to
large will induce excessive wind drag, feet that are to small won't
grasp twigs. Viewed from an IFRG , IOF or IF perspective the feet are
just perfect, made by a perfect Designer(Jesus Christ).

The http://en.wikipedia.org/wiki/Bird-of-paradise could probably be
made more IOF at the expense of its austetics. With the bacterial
flagellum we probably have IOF and none IFRG, in its environment it
has to optimize every single appendage for its function. But with the
fowls, the Lord Jesus made a different statement. Bacterial flagellums
expresses its optimum function by not having as an aestetic appearance
as a titmouse has. We humans as austetic beings want our fowls to be
both functional and pleasing to the eye.

The primary consideration with the B-2 http://en.wikipedia.org/wiki/Stealth_bomber
was low radar signature and not austetics(IOF). With the non-stealth
bombers the designers were more at liberty to work from an IFRG
perspective.
Stealth bombers aren't adapted to their condition of existence but are
described by their attributes. No biological micro or macro machine is
adapted to its condition of existence.

John Harshman

unread,
Jan 27, 2011, 3:11:59 PM1/27/11
to
I would sure like to make some sense of what you say, but you will have
to help out by trying very hard to be coherent. I realize that's unlikely.

When exactly did Darwin claim that tufted titmice had feet that were too
small? I am so far unable to find it in any of Darwin's writings, or in
your reference. It would seem a silly thing for Darwin to say, since
foot size should be easy to change through selection.

Spelling, by the way: you mean "too" and "esthetics", I believe, as in
"feet that are too large" and "at the expense of its esthetics".

Burkhard

unread,
Jan 27, 2011, 3:39:21 PM1/27/11
to

I don;t think he did. Darwin is talking about the feet of the titmouse
in OoS, Under "Objections to the theory of NS sd regarding to
Instincts." He addresses the old "chicken and egg" problem - what good
are beaks better suited for cracking nuts if the bird doesn't like nuts,
and how can a bird like nuts if it can't open them with its beak"

His point is then exactly to say: imagine the bird can already crack
sees,and now its feet are growing, which enables it to climb tress to
get nuts at source - hence forming a new habit.

Nothing about god or inefficient design there.


David Hare-Scott

unread,
Jan 27, 2011, 4:35:59 PM1/27/11
to
backspace wrote:
> http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>
> === Irreducible optimized functionality ===
> The achieve the function or purpose of catching mouse needs five
> parts. There can't be less than five parts or the ''optimum''
> functionality won't be achieved.
>

But to be useful it doesn't need to be optimal nor does it need to be a
functional mouse trap.

> Removing two parts as Kenneth Miller does achieves the fuctionality of
> a powerful paper clip: it no longer can catch mouse automatically.
> Miller is equivocating between different types of functionalities with
> the word "complexity". A better term for Behe's (Irreducbile
> Complexity)IC is '''Irreducible optimized functionality'''. The
> purpose of the mouse trap is to catch mouse ''unattended'' which as
> far as mouse trap engineers can determine needs a minimum of five
> parts. Removing the mouse base and stapling it to the floor doesn't
> remove the ''purpose'' of the original base but replaces one type of
> base for another and it removes the mobility of the mouse trap thus
> reducing its overall purposefulness: Reduced functionality.
>

If you are a human engineer designing mouse traps specifically this is so.
But recall that evolution does not have goals or purpose so already the
analogy is weak.

> Same with the flagellum, it needs 40 parts to swim through the fluid:
> to achieve the purpose of ''Irreducbile Optimized
> Functionality(IOF)''.

You assume that to be useful the flagellum must function as a flagellum.
You assume purpose and hence purposful behaviour. Where there is purposeful
behaviour there is a mind, a powerful mind - oh wait I've just found god!

The type III secretory mechanism which injects
> toxins needs x parts to achieve a different IOF. purpose. They both
> use some of the same shapes, they both need a minimum of 40 and x
> parts respectively in order to '''optimize''' their functionality.
> Some IC systems could still function with less parts but like the
> mouse trap stapled to the ground at the cost of reduced functionality.
>
> With less than 40 parts , a Calvin and Hobb's creation is made, more a
> Rube Goldberg contraption. Both the toxin and flagellum use a round
> shape reducing friction the most. A watch and turbine use round shapes
> because it achives optimized fuctionality.
>

You keep describing these biological "machines" as if they were designed for
a specific purpose. You keep assuming that if they have less than full
function they ought not exist. This is what you are trying to show, you
can't have it as an axiom as well.

David


Mike Painter

unread,
Jan 27, 2011, 5:19:56 PM1/27/11
to
backspace wrote:
> http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>
> === Irreducible optimized functionality ===
> The achieve the function or purpose of catching mouse needs five
> parts. There can't be less than five parts or the ''optimum''
> functionality won't be achieved.

Fiddle. I can build a mouse trap with four parts (three if you don't count
the bait and two if placed in a proper location) that is far better than
your five part model.
The first one I built caught 30 mice in one night and required no attention.

Lessor live trap models have only two parts plus bait.

Not only is your model false but it ignores a huge variety of mouse traps
that have few parts and are more effective.


I do wonder how the number five came up?
1. Spring. (all that is needed in a good design.)
2. bait holder/trigger release.
3. trigger
4 trigger fastener
5 bait holder fastener
6. spring holder
7. base platform.
8. bait

John Harshman

unread,
Jan 27, 2011, 9:02:47 PM1/27/11
to

Actually, he isn't. He calls it a titmouse, but he is in fact referring
to the great tit _Parus major_ rather than the tufted titmouse
_Baeolophus bicolor_.

backspace

unread,
Jan 28, 2011, 1:22:24 AM1/28/11
to
On Jan 27, 11:35 pm, "David Hare-Scott" <sec...@nospam.com> wrote:
> backspace wrote:
> >http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>
> > === Irreducible optimized functionality ===
> > The achieve the function or purpose of catching mouse needs five
> > parts. There can't be less than five parts or the ''optimum''
> > functionality won't be achieved.
>
> But to be useful it doesn't need to be optimal nor does it need to be a
> functional mouse trap.

In other words the concept is over your head?

backspace

unread,
Jan 28, 2011, 1:25:10 AM1/28/11
to
> >>> Thehttp://en.wikipedia.org/wiki/Bird-of-paradisecould probably be

It will be corrected at http://scratchpad.wikia.com/wiki/Irreducible_Functionality
with
http://en.wikipedia.org/wiki/Great_Tit

The article has been edited and the following are proposed to extend
Behe's IC:


1) There will exist biological machines that are purely
IOF(irreducibly optimized functional).

2) There will exist biological machines that combine IF(irreducible
functionality) with Rube Goldberg geniality to make a statement of

austetics: Irreducible functionality Rube Goldberg.

jillery

unread,
Jan 28, 2011, 1:24:39 AM1/28/11
to
> == Austetics ==http://en.wikipedia.org/wiki/Aesthetics. Human designers making

> watches don't always strive for the most optimized solution with
> minimal parts but a combination of Rube Goldberg(RG) and
> optimality(more parts than needed).
> Extending the same notion to IOF three notions are proposed:
> * 1) There will exist biological machines that are purely
> IOF(irreducibly optimized functional).
> * 2) There will exist biological machines that combine IF(irreducible
> functionality) with Rube Goldberg geniality to make a statement of
> austetics: ''Irreducible functionality Rube Goldberg''.
> * 3) The larger the animal the more we tend to have IFRG, as we go
> smaller it will become more IOF until we reach the smallest micro
> machines such as as bacterial flagellum which will be fully IOF and
> not IFRG.
>
> Darwin in one of his passages on thehttp://en.wikipedia.org/wiki/Tufted_Titmouse

> speculated that larger feet would have been a better design for the
> bird. He insinuated that if it was made by God ,that he didn't know
> what he was doing. Instead we should view every single biological
> design as oscillating between IOF and IFRG. The titmouse is optimzed
> for both its functions of grasping twigs and flying: feet that are to
> large will induce excessive wind drag, feet that are to small won't
> grasp twigs. Viewed from an IFRG , IOF or IF perspective the feet are
> just perfect, made by a perfect Designer(Jesus Christ).
>
> Thehttp://en.wikipedia.org/wiki/Bird-of-paradisecould probably be

> made more IOF at the expense of its austetics. With the bacterial
> flagellum we probably have IOF and none IFRG, in its environment it
> has to optimize every single appendage for its function. But with the
> fowls, the Lord Jesus made a different statement. Bacterial flagellums
> expresses its optimum function by not having as an aestetic appearance
> as a titmouse has. We humans as austetic beings want our fowls to be
> both functional and pleasing to the eye.
>
> The primary consideration with the B-2http://en.wikipedia.org/wiki/Stealth_bomber

> was low radar signature and not austetics(IOF). With the non-stealth
> bombers the designers were more at liberty to work from an IFRG
> perspective.
> Stealth bombers aren't adapted to their condition of existence but are
> described by their attributes. No biological micro or macro machine is
> adapted to its condition of existence.

I see at least two problems with your argument here.

One is you insert "optimal" functionality into the definition of IC.
Behe's definition doesn't allow that. It states that the removal of a
part of an IC system makes it non-functional, not less functional. If
after removing a part, the system remains functional at all, no matter
how poorly, it's not IC as far as Behe is concerned.

The other is you ignore Miller's stronger argument against Behe's
IC. Behe says the separate parts must evolve simultaneously, and
that is something evolution can't do. Behe is right that's something
evolution can't do, but it doesn't have to. Instead, the individual
parts of an IC system evolve separately to perform functions different
from what they do in the IC system.

David Hare-Scott

unread,
Jan 28, 2011, 2:01:06 AM1/28/11
to

You are drawing a parallel between designed human artefacts and evolved
living organisms. The analogy has many weaknesses but the one I alluded to
here is that during evolution any change that increases reproductive success
will be selected. The structure or function does not need to be optimal nor
to have some predetermined outcome, as there are no predetermined outcomes,
it just needs to be useful, ie increase reproductive success.

Designing a mousetrap is a different process where the objective is known in
advance if the design doesn't catch mice it is no value as a mousetrap. A
mousetrap with missing parts could well be quite useful (say as a paper
clip) for all that it isn't a mousetrap.

Evolution via selection is nothing like designing mousetraps. Observing
that removing parts from a mousetrap renders it not a mousetrap says nothing
whatsoever about whether complex structures or functions could have evolved
in living organisms.

This is just one flaw in your argument but you have kindly snipped the rest.


David

backspace

unread,
Jan 28, 2011, 2:22:26 AM1/28/11
to
> > Thehttp://en.wikipedia.org/wiki/Bird-of-paradisecouldprobably be

> > made more IOF at the expense of its austetics. With the bacterial
> > flagellum we probably have IOF and none IFRG, in its environment it
> > has to optimize every single appendage for its function. But with the
> > fowls, the Lord Jesus made a different statement. Bacterial flagellums
> > expresses its optimum function by not having as an aestetic appearance
> > as a titmouse has. We humans as austetic beings want our fowls to be
> > both functional and pleasing to the eye.
>
> > The primary consideration with the B-2http://en.wikipedia.org/wiki/Stealth_bomber
> > was low radar signature and not austetics(IOF). With the non-stealth
> > bombers the designers were more at liberty to work from an IFRG
> > perspective.
> > Stealth bombers aren't adapted to their condition of existence but are
> > described by their attributes. No biological micro or macro machine is
> > adapted to its condition of existence.
>
> I see at least two problems with your argument here.
>
> One is you insert "optimal" functionality into the definition of IC.
> Behe's definition doesn't allow that. It states that the removal of a
> part of an IC system makes it non-functional, not less functional. If
> after removing a part, the system remains functional at all, no matter
> how poorly, it's not IC as far as Behe is concerned.

Indeed this was Behe's initial formulation. We must not gravitate on
initial views but extend such views. Nobody has an intellectual
monopoly on the concept of
IC . The ideas isn't original to Behe. For example Popper's idea of
falsifiability was foreshadowed by Charles Hodge.
http://scratchpad.wikia.com/wiki/Charles_Hodge
http://scratchpad.wikia.com/wiki/Popper

> Behe says the separate parts must evolve simultaneously, and
> that is something evolution can't do.

Evolution in the pattern or design sense?

backspace

unread,
Jan 28, 2011, 2:45:12 AM1/28/11
to
On Jan 28, 9:01 am, "David Hare-Scott" <sec...@nospam.com> wrote:
> backspace wrote:
> > On Jan 27, 11:35 pm, "David Hare-Scott" <sec...@nospam.com> wrote:
> >> backspace wrote:
> >>>http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>
> >>> === Irreducible optimized functionality ===
> >>> The achieve the function or purpose of catching mouse needs five
> >>> parts. There can't be less than five parts or the ''optimum''
> >>> functionality won't be achieved.
>
> >> But to be useful it doesn't need to be optimal nor does it need to
> >> be a functional mouse trap.
>
> > In other words the concept is over your head?
>
> You are drawing a parallel between designed human artefacts and evolved
> living organisms.  The analogy has many weaknesses but the one I alluded to
> here is that during evolution any change that increases reproductive success
> will be selected.  The structure or function does not need to be optimal nor
> to have some predetermined outcome, as there are no predetermined outcomes,
> it just needs to be useful, ie increase reproductive success.

success, useful and selection all refer to the same fact making your
paragraph a rhetorical tautology: the general thrust of it is stated
so that it can't be disputed.


> Designing a mousetrap is a different process where the objective is known in
> advance if the design doesn't catch mice it is no value as a mousetrap.

Problem comes with the word evolution and what we mean with it in the
pattern or design sense.


> A
> mousetrap with missing parts could well be quite useful (say as a paper
> clip) for all that it isn't a mousetrap.

The purpose is to catch mouse unattended which as far as anybody can
see needs at least five parts all working at the same time.

> Evolution via selection is nothing like designing mousetraps.

Who did the selecting?

> Observing that removing parts from a mousetrap renders it not a mousetrap says nothing
> whatsoever about whether complex structures or functions could have evolved
> in living organisms.

Correct because you must define what you mean with evolution: pattern
or design .


backspace

unread,
Jan 28, 2011, 6:16:07 AM1/28/11
to
On Jan 28, 12:19 am, "Mike Painter" <md.pain...@sbcglobal.net> wrote:
> backspace wrote:
> >http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>
> > === Irreducible optimized functionality ===
> > The achieve the function or purpose of catching mouse needs five
> > parts. There can't be less than five parts or the ''optimum''
> > functionality won't be achieved.
>
> Fiddle. I can build a mouse trap with four parts (three if you don't count
> the bait and two if placed in a proper location) that is far better than
> your five part model.
> The first one I built caught 30 mice in one night and required no attention.

That might be possible but you are missing the point of the analogy.
If you would describe your mouse trap we would perhaps discover that
having three parts make its less optimal in terms of mobility. It all
depends what the designer set out to do. A factory manager producing
thousands of traps needs to use the least parts for the most optimized
function, yet still be able to sell the device. The intent of the
designer is crucial which is why I extended IC to IF, IOF and IFRG.

> Lessor live trap models have only two parts plus bait.

Are they mass producible and mobile? What would be the intent of the
trap. Are they to be sold .


> Not only is your model false but it ignores a huge variety of mouse traps
> that have few parts and are more effective.

Effectiveness is depended on the financial , material and engineering
constraints. Nothing is being ignored , functionality
is depended on the engineering problem at hand within certain
constraints. Flagellum's have different constraints that titmouse
birds but neither the bird nor flagella are adapted to their
environment.


> I do wonder how the number five came up?
> 1. Spring. (all that is needed in a good design.)
> 2. bait holder/trigger release.
> 3. trigger
> 4 trigger fastener
> 5 bait holder fastener
> 6. spring holder
> 7. base platform.
> 8. bait

Nr.8 is a good point of yours. If we want to catch mouse we obviously
need bait! I didn't think about this, thanks for the insigth. Thus at
least 6 parts would be needed to achieve a stated function. But the
factory owner only needs to produce five with the client adding the
sixth part. We could probably carry on like this forever and miss the
essential point that the analogy is trying to convey.

Burkhard

unread,
Jan 28, 2011, 6:27:00 AM1/28/11
to
On 27/01/2011 20:03, backspace wrote:
> http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>
> === Irreducible optimized functionality ===
> The achieve the function or purpose of catching mouse needs five
> parts. There can't be less than five parts or the ''optimum''
> functionality won't be achieved.
>
> Removing two parts as Kenneth Miller does achieves the fuctionality of
> a powerful paper clip: it no longer can catch mouse automatically.
> Miller is equivocating between different types of functionalities with
> the word "complexity". A better term for Behe's (Irreducbile
> Complexity)IC is '''Irreducible optimized functionality'''.


I'd say it is rather a totally different term. Which is fine, only it
cannot play any longer the role Behe had intended for it - that is,
nothing of particular interest would follows from it if we were to
observe a organism having "irreducible optimized functionality".

jillery

unread,
Jan 28, 2011, 8:41:12 AM1/28/11
to
> falsifiability was foreshadowed by Charles Hodge:

http://scratchpad.wikia.com/wiki/Charles_Hodgehttp://scratchpad.wikia.com/wiki/Popper

I mention Behe only because it's his definition Behe and Miller use
when discussing Behe's IC mousetrap. It's disingenuous for you to
disagree with Miller by substituting your own definition.

More to the point, "optimal" is a variable that depends on what's
being measured. For example optimizing for thermal efficiency, speed,
and yield may require three entirely different designs. Adding
"optimal" to IC adds more smoke than light.

> >  Behe says the separate parts must evolve simultaneously, and
> > that is something evolution can't do.
>
> Evolution in the pattern or design sense?

In whatever sense Behe and Miller used.

backspace

unread,
Jan 28, 2011, 10:22:35 AM1/28/11
to
> http://scratchpad.wikia.com/wiki/Charles_Hodgehttp://scratchpad.wikia...

>
> I mention Behe only because it's his definition Behe and Miller use
> when discussing Behe's IC mousetrap.  It's disingenuous for you to
> disagree with Miller by substituting your own definition.

Miller obfuscates Behe's intent. His analogy isn't perfect and neither
is mine. Miller commits:
1) Strawman arguments.
2) Out of context red herrings.

In addition the grammatical gargoyle natural selection which I used in
the ''preferential descision'' sense in another thread isn't helping
us gaining insight into biology as a living machine. It is like
imagine we are poking partial differential equations into a computer
trying to model an object and somebody enters a single rote same
universal differential equation every 20th lien: Our calculations will
jam.

> More to the point, "optimal" is a variable that depends on what's
> being measured.  For example optimizing for thermal efficiency, speed,
> and yield may require three entirely different designs.  Adding
> "optimal" to IC adds more smoke than light.

I have a right to my ideas and can use any term and extend any
concept.You are free to refute or reject them.


> > >  Behe says the separate parts must evolve simultaneously, and
> > > that is something evolution can't do.
>
> > Evolution in the pattern or design sense?

> In whatever sense Behe and Miller used.

Which is ?


Erwin Moller

unread,
Jan 28, 2011, 10:41:29 AM1/28/11
to
On 1/28/2011 8:45 AM, backspace wrote:
> On Jan 28, 9:01 am, "David Hare-Scott"<sec...@nospam.com> wrote:
>> backspace wrote:
>>> On Jan 27, 11:35 pm, "David Hare-Scott"<sec...@nospam.com> wrote:
>>>> backspace wrote:
>>>>> http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>>
>>>>> === Irreducible optimized functionality ===
>>>>> The achieve the function or purpose of catching mouse needs five
>>>>> parts. There can't be less than five parts or the ''optimum''
>>>>> functionality won't be achieved.
>>
>>>> But to be useful it doesn't need to be optimal nor does it need to
>>>> be a functional mouse trap.
>>
>>> In other words the concept is over your head?
>>
>> You are drawing a parallel between designed human artefacts and evolved
>> living organisms. The analogy has many weaknesses but the one I alluded to
>> here is that during evolution any change that increases reproductive success
>> will be selected. The structure or function does not need to be optimal nor
>> to have some predetermined outcome, as there are no predetermined outcomes,
>> it just needs to be useful, ie increase reproductive success.
>
> success, useful and selection all refer to the same fact making your
> paragraph a rhetorical tautology: the general thrust of it is stated
> so that it can't be disputed.
>

Aaah, at last, the tautology!
No Backspace posting is complete without it. ;-)

Still trying to disprove evolution with a dictionary?


>
>> Designing a mousetrap is a different process where the objective is known in
>> advance if the design doesn't catch mice it is no value as a mousetrap.
>
> Problem comes with the word evolution and what we mean with it in the
> pattern or design sense.

So, in short, you simply refuse to address David's comments.
Let me guess why: Evolution is a tautology, so why address Davids valid
observation that purposeful intelligent design (an engineered mousetrap)
is something else than evolution?


>> A
>> mousetrap with missing parts could well be quite useful (say as a paper
>> clip) for all that it isn't a mousetrap.
>
> The purpose is to catch mouse unattended which as far as anybody can
> see needs at least five parts all working at the same time.

Yes, we know that.
But you are missing that some parts of that very mousetrap could serve a
purpose DIFFERENT from trapping mice.

>
>> Evolution via selection is nothing like designing mousetraps.
>
> Who did the selecting?

Who pulls the rock onto the Earth?
Why doesn't it float?
Who is pulling that thing?


The critters that survive, survive.
The critter that leave offspring, leave offspring, hence spread their genes.
The critter that die before making small critters, are goners, and don't
spread their genes.

So in case you insist in getting an answer to "Who did the selecting?"
it would be: Nobody. Or maybe the universe (if that qualifies as a "who".)
A better question would be: "What causes the selecting?"

It is not so complicated once you try to understand the mechanism
instead of trying to understand your dictionary and make reality fit
what-ever it is you found in there.


>
>> Observing that removing parts from a mousetrap renders it not a mousetrap says nothing
>> whatsoever about whether complex structures or functions could have evolved
>> in living organisms.
>
> Correct because you must define what you mean with evolution: pattern
> or design .

pattern or design?
Coffee or milk?

What do you mean?

Regards,
Erwin Moller

--
"That which can be asserted without evidence, can be dismissed without
evidence."
-- Christopher Hitchens

John Harshman

unread,
Jan 28, 2011, 10:51:07 AM1/28/11
to
That's certainly opaque. Is unintelligibility what you were aiming for?

Mike Lyle

unread,
Jan 28, 2011, 11:15:52 AM1/28/11
to
On Fri, 28 Jan 2011 07:22:35 -0800 (PST), backspace
<steph...@gmail.com> wrote:

>On Jan 28, 3:41�pm, jillery <69jpi...@gmail.com> wrote:

[...]


>
>> More to the point, "optimal" is a variable that depends on what's
>> being measured. �For example optimizing for thermal efficiency, speed,
>> and yield may require three entirely different designs. �Adding
>> "optimal" to IC adds more smoke than light.
>
>I have a right to my ideas and can use any term and extend any
>concept.You are free to refute or reject them.
>

Insofar as a discourse in terms of "rights" is appropriate in this
case, you have a legal right of sorts, and probably a moral right, to
your ideas. You don't, however, have an intellectual right to them
until you've earned it by showing them to be reasonable.

Perhaps similarly, language is a cooperative venture: you have no
meaningful or useful "right" to insist that others must accept
word-meanings on which there is no consensus. If you want people to
discuss things with you, you must use generally-accepted language.
>
>> > > �Behe says the separate parts must evolve simultaneously, and


>> > > that is something evolution can't do.
>>
>> > Evolution in the pattern or design sense?

Please don't start that all over again.
[...]
--
Mike.

hersheyh

unread,
Jan 28, 2011, 12:49:38 PM1/28/11
to
> >>>>> Thehttp://en.wikipedia.org/wiki/Bird-of-paradisecouldprobably be
> > It will be corrected athttp://scratchpad.wikia.com/wiki/Irreducible_Functionality

> > with
> >http://en.wikipedia.org/wiki/Great_Tit
>
> > The article has been edited and the following are proposed to extend
> > Behe's IC:
> > 1) There will exist biological machines that are purely
> > IOF(irreducibly optimized functional).
> > 2) There will exist biological machines that combine IF(irreducible
> > functionality) with Rube Goldberg geniality to make a statement of
> > austetics: Irreducible functionality Rube Goldberg.
> > 3) The larger the animal the more we tend to have IFRG, as we go
> > smaller it will become more IOF until we reach the smallest micro
> > machines such as as bacterial flagellum which will be fully IOF and
> > not IFRG.
>
> That's certainly opaque. Is unintelligibility what you were aiming for?

I confess I saw more than a passing resemblance between backspace and
another person who has a strong interest in "logic" (but also one
whose reasoning only makes sense to himself) and who thought that "The
government (scientists, in backspace's private alternate dream
reality) is implying mind control and brainwash on the people by
controlling grammar.� backspace also seems to believe in a �truth
language� that has been discarded. Even mathematics becomes relative
and grammatically controlled in such a fantasy world. �My instructor
said he called a number 6, and I said I call it 18." �How can you
deny math instead of accept it?�

http://www.newsweek.com/2011/01/10/jared-lee-loughner-s-mental-state.html

I have, of course, no reason at all to think that backspace (unlike
the above person) is of any danger to others. Most (the vast, vast
majority) people with distorted thinking are not and many can even be
productive citizens. And, unlike the above person, I feel no hint of
"anger" in backspace's responses to others here. But I do admit that
he was one of the first people I thought about as having similarity in
"thought" processes. I am not a psychologist, nor do I even wish to
play one on the net, so take that into consideration. Just pointing
out my impressions of some similarities in thought processes that
occurred to me at the time.

johnetho...@yahoo.com

unread,
Jan 28, 2011, 2:31:12 PM1/28/11
to
On Jan 28, 8:15�am, Mike Lyle <mike_lyle...@yahoo.co.uk> wrote:
> On Fri, 28 Jan 2011 07:22:35 -0800 (PST), backspace

<snip various backspace idiocies and people's attempts to give him
clue>

>
> Please don't start that all over again.
> [...]
> --
> Mike

People have been telling children not to masturbate in public for ages
but it has no effect until the child develops certain a minimal level
of maturity. I don't expect backspace will ever stop.

backspace

unread,
Jan 28, 2011, 6:28:46 PM1/28/11
to
On Jan 28, 8:49 pm, hersheyh <hershe...@yahoo.com> wrote:
> controlling grammar.  backspace also seems to believe in a truth
> language that has been discarded.  Even mathematics becomes relative
> and grammatically controlled in such a fantasy world.   My instructor
> said he called a number 6, and I said I call it 18."   How can you

> deny math instead of accept it?
>
> http://www.newsweek.com/2011/01/10/jared-lee-loughner-s-mental-state....

>
> I have, of course, no reason at all to think that backspace (unlike
> the above person) is of any danger to others.  Most (the vast, vast
> majority) people with distorted thinking are not and many can even be
> productive citizens.  And, unlike the above person, I feel no hint of
> "anger" in backspace's responses to others here.  But I do admit that
> he was one of the first people I thought about as having similarity in
> "thought" processes.  I am not a psychologist, nor do I even wish to
> play one on the net, so take that into consideration.  Just pointing
> out my impressions of some similarities in thought processes that
> occurred to me at the time.

Our political and social discourse have embalmed into hyperbole. From
Glen Beck's acidifying oratory to the cornification of America by
Sarah Palin and emotional slop from Oprah Winfrey the American mind
has entered a type of intimate cartoonish universe, its sense of the
spiritual and eternal oxidized.

A most unfortunate event transpired in that a mother and wife was shot
by what seems to be a paranoid schizophrenic. With cunning and callous
precision this was used to target political opponents for ratings. A
schizophrenic can be influenced by over the top metaphorical imagery
and statements, uttered in a context where acting like a Howler monkey
is preferred above the serenity of a contemplating logician.

The Democrats and Obama were first to intone that the ''political
rhetoric'' must be scaled down. Presumably Obama isn't engaging
himself in any gladiatorial battles of demagoguery but in rational
engagement,insinuating that the Republicans in general and Sarah Palin
in particular are responsible for a schizophrenic out of touch with
reality.

Robert Camp

unread,
Jan 28, 2011, 9:01:48 PM1/28/11
to

"Acidifying," "cornification," "oxidized.?" I like a good turn of
phrase as much as the next guy, but do you actually have any idea what
you're trying to say?

> A most unfortunate event transpired in that a mother and wife was shot
> by what seems to be a paranoid schizophrenic. With cunning and callous
> precision this was used to target �political opponents for ratings. A
> schizophrenic can be influenced by over the top metaphorical imagery
> and statements, uttered in a context where acting like a Howler monkey
> is preferred above the serenity of a contemplating logician.

Oh, I get it. You're trying to say what you've always tried to say -
"You're all nuts, I'm the only one who's sane."

How's that bit of myopia working out so far?

> The Democrats and Obama were first to intone that the ''political
> rhetoric'' must be scaled down. Presumably Obama isn't engaging
> himself in any gladiatorial battles of demagoguery but in rational
> engagement,insinuating that the Republicans in general and Sarah Palin
> in particular are responsible for a schizophrenic out of touch with
> reality.

(Following your derailed train of thought off the cliff...)

I don't know who was first to "intone" (did someone just give you your
first thesaurus for Christmas?) but people, including the woman who
was the object of the attack, were suggesting the rhetoric was getting
out of hand before the shooting in Tucson, so this isn't just
opportunism, this is a continuation of conscience.

Now of course there's no way - short of Loughner himself admitting to
being so influenced - to create a direct causal link between foolish
and flagrant verbal imagery and metaphor, and the actions of a madman.
But that's not the point. It's about whether the loose talk leads to a
social and political context in which the loonies find motivation.

Consider an analogy. Although Hurricane Katrina cannot be directly
linked to global warming, the current best information suggests that
rising sea temperatures create conditions that lead to stronger
storms. Similarly, I don't think Palin was responsible for Loughner,
but she may have played a part in creating a public climate in which a
nut-job like him felt encouraged to act out.

In other words, it doesn't do anyone any good to assign blame (and I
agree with you that some pundits and elected officials have come close
to doing so), but neither is it reasonable to ignore those possible
contributing factors that might help inform our efforts to keep this
from happening again.

Maybe if we lower the temperature just a bit, we can keep this kind of
storm a rarity.

RLC

David Hare-Scott

unread,
Jan 29, 2011, 1:44:18 AM1/29/11
to

Oh groan that same bloody treadmill is still turning in your head. You try
to play this word game every few months and it always leads the same place,
precisely nowhere. But that is what treadmills are designed to do.

Also I find your practice of dismembering my posts down to meaningless sound
bites and then attaching your empty word-game dogma to the fragment is not
only stupid but rude. To insist on replying to context-free fragments
instead of what I actually typed is hijacking the interchange far beyond
reasonable snippage. Go away and stop wasting my time.

David


BramH

unread,
Jan 29, 2011, 4:27:00 AM1/29/11
to
> >  Behe says the separate parts must evolve simultaneously, and
> > that is something evolution can't do.
>

Consider gazelles being chased by cheetahs ( i get this example from
one of Ernst Mayr's books.)
Try to deny any of the following:

1 In the gazelles, because they are hunted, there is selection for
running capabilities.

2 So there is selection for longer legs and Gazelles will evolve
longer legs

3 However Gazelles have to eat.

4 In order to eat, they need a long neck to reach the ground.

5 If they evolve longer legs without evolving a longer neck they will
starve.

6 Meaning that, in gazelles with longer legs, there will be
differential survival due to neck-length.

7 This will cause the evolution of longer necks.

The important thing to notice is that the evolution of longer legs
will initiate the selection for longer necks and hence cause the
evolution of longer necks.

Coclusion: because organisms consist of integrated functional wholes,
parts must evolve necessarily together.

backspace

unread,
Jan 29, 2011, 9:05:17 AM1/29/11
to
On Jan 29, 11:27�am, BramH <br.hess...@planet.nl> wrote:
> > > �Behe says the separate parts must evolve simultaneously, and

> > > that is something evolution can't do.

> Consider gazelles being chased by cheetahs ( i get this example from
> one of Ernst Mayr's books.)

> Try to deny any of the following:

Right here is your problem. In the immortal words of Darwin himself
''... the truth of the propositions cannot be disputed...'' Popper
demonstrated that if your statements or theory can't be disputed(if
there is no way to test them) they aren't theories but logical
fallacies. A theory is defined as something which can be disputed. We
are after a theory of derivation(Prof. Owen's Doctrine of Derivation)
not a series of truisms from which follows a non-sequitur.
We are told that due to Uranium decay the earth is 5bil years old.
Then the Uranium is dated to 5bil years due to the earth being 5bil
years old. This is circular reasoning, the earth is dated to 5bil
using UR and then UR dated to 5bil using the earth. What we need is an
independant means of dating the UR itself. Because we don't know how
UR came into existance.

> 4 In order to eat, they need a long neck to reach the ground.

As you said nothing can be disputed.

> 7 This will cause the evolution of longer necks.

Non-sequitur.

TomS

unread,
Jan 29, 2011, 9:27:07 AM1/29/11
to
"On Sat, 29 Jan 2011 01:27:00 -0800 (PST), in article
<c42a3fab-53e4-4191...@e13g2000yqb.googlegroups.com>, BramH
stated..."
>
>> > �Behe says the separate parts must evolve simultaneously, and
Spencer used the example of the neck muscles of the Irish Elk. See
the references in the Wikipedia article on "Irreducible complexity"
under forerunners from the nineteenth century.


--
---Tom S.
"... the heavy people know some magic that can make things move and even fly,
but they're not very bright, because they can't survive without their magic
contrivances"
Xixo, in "The Gods Must Be Crazy II"

jillery

unread,
Jan 29, 2011, 10:14:57 AM1/29/11
to

I believe you honestly misunderstand the context. I paraphrase Behe's
argument for ID using IC as follows:

1. I observe complex biological systems with parts, and each part is
necessary for the function of the system. By my definition, they are
Irreducibly Complex.
2. To evolve IC systems, Natural Selection must select variations
based on what that part will do in the future, after the IC systems
are put together.
3. Since NS has no foresight, and can only select variations using
present conditions, therefore NS, and so evolution, can't explain IC
systems.
4. IC proves an intelligent designer. QED

Your analogy is about separate parts evolving in parallel in response
to the immediate environment, which includes the changing demands
wrought from other changing parts. To apply Behe's argument to your
analogy, the parts of the gazelle evolve in response to an
intelligent designer anticipating the future needs for longer legs and
necks. Hopefully, you can see the difference.

backspace

unread,
Jan 29, 2011, 1:46:09 PM1/29/11
to
On Jan 29, 5:14 pm, jillery <69jpi...@gmail.com> wrote:
> 3. Since NS has no foresight, and can only select variations using
> present conditions, therefore NS, and so evolution, can't explain IC
> systems.

When you say NS has no ''foresight'' , do you mean the concept NS1 or
NS4?

The concept represented with NS spans the time era from 1831(Patrick
Matthews) to 2011 a period of 180 years.
We have:
1) NS1 - 1831 - 1874A.D which had nothing to do with Genes or IC
systems. It represented the Malthus theory.
2) NS2 - Genes a physical molecule.
3) NS3 - Genes as cybernetic abstraction, represents digital
information.
4) NS4 - 2010A.D used in the *preferential decision* sense in the
talk.origins usenet thread on Automated Selection.

http://newsgroups.derkeiler.com/Archive/Talk/talk.origins/2010-09/msg01337.html

Some would object and say that 'natural selection' doesn't mean
'preferential decision' which is correct since NS isn't even a
sentence and Sentences Have No Meaning and neither words, thus neither
can a term like NS have any meaning. My idea though as a YEC in terms
of the pattern/design, disorder/order, cause/effect sense has meaning.
Pattern, design, selection and natural are but tools to convey such
meaning within a knowledge context of genes as a cybernetic
abstraction.

Bob Casanova

unread,
Jan 29, 2011, 4:18:33 PM1/29/11
to
On Sat, 29 Jan 2011 10:46:09 -0800 (PST), the following
appeared in talk.origins, posted by backspace
<steph...@gmail.com>:

>On Jan 29, 5:14 pm, jillery <69jpi...@gmail.com> wrote:
>> 3. Since NS has no foresight, and can only select variations using
>> present conditions, therefore NS, and so evolution, can't explain IC
>> systems.
>
>When you say NS has no ''foresight'' , do you mean the concept NS1 or
>NS4?

He means that selection acts without intent, on current
populations in current environments. As you know very well.

>The concept represented with NS spans the time era from 1831(Patrick
>Matthews) to 2011 a period of 180 years.
>We have:
>1) NS1 - 1831 - 1874A.D which had nothing to do with Genes or IC
>systems. It represented the Malthus theory.
>2) NS2 - Genes a physical molecule.
>3) NS3 - Genes as cybernetic abstraction, represents digital
>information.
>4) NS4 - 2010A.D used in the *preferential decision* sense in the
>talk.origins usenet thread on Automated Selection.
>
>http://newsgroups.derkeiler.com/Archive/Talk/talk.origins/2010-09/msg01337.html
>
>Some would object and say that 'natural selection' doesn't mean
>'preferential decision' which is correct since NS isn't even a
>sentence and Sentences Have No Meaning and neither words, thus neither
>can a term like NS have any meaning. My idea though as a YEC in terms
>of the pattern/design, disorder/order, cause/effect sense has meaning.
>Pattern, design, selection and natural are but tools to convey such
>meaning within a knowledge context of genes as a cybernetic
>abstraction.

Would you prefer balsamic vinaigrette or Thousand Island on
that?
--

Bob C.

"Evidence confirming an observation is
evidence that the observation is wrong."
- McNameless

Mike Lyle

unread,
Jan 29, 2011, 4:36:10 PM1/29/11
to
On Sat, 29 Jan 2011 14:18:33 -0700, Bob Casanova <nos...@buzz.off>
wrote:

Michelin star for the best self-parody I've seen in maybe years,
though. Any bets on whether he goes for the second star, or rests on
his bay-leaves?

--
Mike.

backspace

unread,
Jan 29, 2011, 4:39:40 PM1/29/11
to
On Jan 29, 11:18 pm, Bob Casanova <nos...@buzz.off> wrote:
> On Sat, 29 Jan 2011 10:46:09 -0800 (PST), the following
> appeared in talk.origins, posted by backspace
> <stephan...@gmail.com>:

>
> >On Jan 29, 5:14 pm, jillery <69jpi...@gmail.com> wrote:
> >> 3. Since NS has no foresight, and can only select variations using
> >> present conditions, therefore NS, and so evolution, can't explain IC
> >> systems.
>
> >When you say NS has no ''foresight'' , do you mean the concept NS1 or
> >NS4?
>
> He means that selection acts without intent, on current
> populations in current environments. As you know very well.

One can enact a selection for cafe over banana. But to say
"....selection acts without intent....." is grammatically corerct but
as meaningless as Colorless green ideas sleep furiously. The entire
origins debate is bogged down with meaningless sentences.

1) At http://bit.ly/gVwUAQ the poster wrote "...To evolve IC systems,


Natural Selection must select variations based on what that part will

do in the future, after the IC systems are put together....."

Lets rephrase this in terms of PD.
2) "...To evolve IC systems, Preferential Decision(Natural Selection)


must select variations based on what that part will do in the future,

after the IC systems are put together....."

Both Sentence 1) and 2) are grammatically correct but just as
meaningless as
http://en.wikipedia.org/wiki/Colorless_green_ideas_sleep_furiously

Thus lets rephrase again for one possible but not neccassarily correct
idea(It depends what the user of the objects decision, natural,
selection is trying to convey. Whether his concept is correct or not
or conflicts with your world view is irrelevant. In the context of
this section what he actually intends is the issue not whether one
might agree with it or not):

Lets rephrase this in terms of PD or NS:
3) "...To evolve IC systems, the Gaia nature selection force enacts a
Preferential Decision(Natural Selection) for the variant based on what
that part will do in the future, after the IC system is assembled by
Lord Gaia ....."

Most users of the object NS won't intend 3) but all would agree that
with 3) we can deduce the exact meaning the sentence is both
grammatically correct and meaningful. Not so though with 1) and 2)

Conclusion: The way NS and DRS are used in sentences is grammatically
correct but meaningless. They are not "wrong", it is far worse than
being wrong but meaningless. If only we could say something is right
or wrong but the grammatically correct sentence constructions are
cleverly disguised undefined deceit : Until it is defined what is
meant, the concept like a colorless green idea isn't even wrong.

jillery

unread,
Jan 29, 2011, 4:50:23 PM1/29/11
to
On Jan 29, 1:46�pm, backspace <stephan...@gmail.com> wrote:
> On Jan 29, 5:14�pm, jillery <69jpi...@gmail.com> wrote:
>
> > 3. Since NS has no foresight, and can only select variations using
> > present conditions, therefore NS, and so evolution, can't explain IC
> > systems.
>
> When you say NS has no ''foresight'' , do you mean the concept NS1 or
> NS4?
>
> The concept represented with NS spans the time era from 1831(Patrick
> Matthews) to 2011 a period of 180 years.
> We have:
> 1) NS1 - �1831 - 1874A.D which had nothing to do with Genes or IC
> systems. It represented the Malthus theory.
> 2) NS2 - Genes a physical molecule.
> 3) NS3 - �Genes as cybernetic abstraction, represents digital
> information.
> 4) NS4 - 2010A.D �used in the *preferential decision* sense in the
> talk.origins usenet thread on Automated Selection.
>
> http://newsgroups.derkeiler.com/Archive/Talk/talk.origins/2010-09/msg...

>
> Some would object and say that 'natural selection' doesn't mean
> 'preferential decision' which is correct since NS isn't even a
> sentence and Sentences Have No Meaning and neither words, thus neither
> can a term like NS have any meaning. My idea though as a YEC in terms
> of the pattern/design, disorder/order, cause/effect sense has meaning.
> Pattern, design, selection and natural are but tools to convey such
> meaning within a knowledge context of genes as a cybernetic
> abstraction.

If words have no meaning, then none of us can understand what anybody
else writes. So your question is as meaningless to me as my answer is
to you, and there's no point in posting either.

Burkhard

unread,
Jan 29, 2011, 4:51:41 PM1/29/11
to
On Jan 29, 2:05�pm, backspace <stephan...@gmail.com> wrote:
> On Jan 29, 11:27 am, BramH <br.hess...@planet.nl> wrote:
>
> > > > Behe says the separate parts must evolve simultaneously, and
> > > > that is something evolution can't do.
> > Consider gazelles being chased by cheetahs ( i get this example from
> > one of Ernst Mayr's books.)
> > Try to deny any of the following:
>
> Right here is your problem. In the immortal words of Darwin himself
> ''... the truth of the propositions cannot be disputed...'' Popper
> demonstrated that if your statements or theory can't be disputed(if
> there is no way to test them) they aren't theories but logical
> fallacies.

Only that the context of the quote makes it very clear that what
Darwin means is simply: "there is so strong positive evidence for this
claim that for the purpose of this argument, we can assume it is
true"� This has nothing to do with the possibility in principle to
falsify it, just that the bruden would be on any one raisingt the
doubt.

David Hare-Scott

unread,
Jan 29, 2011, 4:57:43 PM1/29/11
to

But adding "optimal" in as another constraint is handy when you aiming to
find The Designer. The more characteristics of design you can get
incorporated into the analogy the more chance it will magically demonstrate
there is a designer.

>>> Behe says the separate parts must evolve simultaneously, and
>>> that is something evolution can't do.
>>
>> Evolution in the pattern or design sense?
>
> In whatever sense Behe and Miller used.

This game over senses of some key words is the whole reason for starting
this and every thread. Don't go in there unless you want to see all your
words sucked inexorably into the black hole of meaning at the centre of
backspace's mind.

David

jillery

unread,
Jan 29, 2011, 4:57:19 PM1/29/11
to
On Jan 29, 4:39�pm, backspace <stephan...@gmail.com> wrote:
> On Jan 29, 11:18�pm, Bob Casanova <nos...@buzz.off> wrote:
>
> > On Sat, 29 Jan 2011 10:46:09 -0800 (PST), the following
> > appeared in talk.origins, posted by backspace
> > <stephan...@gmail.com>:
>
> > >On Jan 29, 5:14�pm, jillery <69jpi...@gmail.com> wrote:
> > >> 3. Since NS has no foresight, and can only select variations using
> > >> present conditions, therefore NS, and so evolution, can't explain IC
> > >> systems.
>
> > >When you say NS has no ''foresight'' , do you mean the concept NS1 or
> > >NS4?
>
> > He means that selection acts without intent, on current
> > populations in current environments. As you know very well.
>
> One can enact a selection for cafe over banana. But to say
> "....selection acts without intent....." is grammatically corerct but
> as meaningless as Colorless green ideas sleep furiously. The entire
> origins debate is bogged down with meaningless sentences.
>
> 1) Athttp://bit.ly/gVwUAQthe poster wrote "...To evolve IC systems,

> Natural Selection must select variations based on what that part will
> do in the future, after the IC systems are put together....."
>
> Lets rephrase this in terms of PD.
> 2) "...To evolve IC systems, Preferential Decision(Natural Selection)
> must select variations based on what that part will do in the future,
> after the IC systems are put together....."
>
> Both Sentence 1) and 2) are grammatically correct but just as
> meaningless ashttp://en.wikipedia.org/wiki/Colorless_green_ideas_sleep_furiously

>
> Thus lets rephrase again for one possible but not neccassarily correct
> idea(It depends what the user of the objects decision, natural,
> selection is trying to convey. Whether his concept is correct or not
> or conflicts with your world view is irrelevant. In the context of
> this section what he actually intends is the issue not whether one
> might agree with it or not):
>
> Lets rephrase this in terms of PD or NS:
> 3) "...To evolve IC systems, the Gaia nature selection force enacts a
> Preferential Decision(Natural Selection) for the variant based on what
> that part will do in the future, after the IC system is assembled by
> Lord Gaia ....."
>
> Most users of the object NS won't intend 3) but all would agree that
> with 3) we can deduce the exact meaning the sentence is both
> grammatically correct and meaningful. Not so though with 1) and 2)
>
> Conclusion: The way NS and DRS are used in sentences is grammatically
> correct but meaningless. They are not "wrong", it is far worse than
> being wrong but meaningless. If only we could say something is right
> or wrong but the grammatically correct sentence constructions are
> cleverly disguised undefined deceit : Until it is defined what is
> meant, the concept like a colorless green idea isn't even wrong.

My impression is you are trying to single-handedly prove your
assertion that "Sentences Have No Meaning and neither words". You
certainly convinced me in your case.

jillery

unread,
Jan 29, 2011, 5:17:49 PM1/29/11
to
> >>>> thehttp://en.wikipedia.org/wiki/Tufted_Titmousespeculated that

> >>>> larger feet would have been a better design for the
> >>>> bird. He insinuated that if it was made by God ,that he didn't know
> >>>> what he was doing. Instead we should view every single biological
> >>>> design as oscillating between IOF and IFRG. The titmouse is
> >>>> optimzed
> >>>> for both its functions of grasping twigs and flying: feet that are
> >>>> to
> >>>> large will induce excessive wind drag, feet that are to small won't
> >>>> grasp twigs. Viewed from an IFRG , IOF or IF perspective the feet
> >>>> are
> >>>> just perfect, made by a perfect Designer(Jesus Christ).
>
> >>>> Thehttp://en.wikipedia.org/wiki/Bird-of-paradisecouldprobablybe
> >>>> made more IOF at the expense of its austetics. With the bacterial
> >>>> flagellum we probably have IOF and none IFRG, in its environment it
> >>>> has to optimize every single appendage for its function. But with
> >>>> the
> >>>> fowls, the Lord Jesus made a different statement. Bacterial
> >>>> flagellums
> >>>> expresses its optimum function by not having as an aestetic
> >>>> appearance
> >>>> as a titmouse has. We humans as austetic beings want our fowls to
> >>>> be
> >>>> both functional and pleasing to the eye.
>
> >>>> The primary consideration with the
> >>>> B-2http://en.wikipedia.org/wiki/Stealth_bomberwas low radar

> >>>> signature and not austetics(IOF). With the non-stealth
> >>>> bombers the designers were more at liberty to work from an IFRG
> >>>> perspective.
> >>>> Stealth bombers aren't adapted to their condition of existence but
> >>>> are
> >>>> described by their attributes. No biological micro or macro
> >>>> machine is
> >>>> adapted to its condition of existence.
>
> >>> I see at least two problems with your argument here.
>
> >>> One is you insert "optimal" functionality into the definition of IC.
> >>> Behe's definition doesn't allow that. It states that the removal of
> >>> a part of an IC system makes it non-functional, not less
> >>> functional. If after removing a part, the system remains functional
> >>> at all, no matter how poorly, it's not IC as far as Behe is
> >>> concerned.
>
> >> Indeed this was Behe's initial formulation. We must not gravitate on
> >> initial views but extend such views. Nobody has an intellectual
> >> monopoly on the concept of
> >> IC . The ideas isn't original to Behe. For example Popper's idea of
> >> falsifiability was foreshadowed by Charles Hodge:
>
> >http://scratchpad.wikia.com/wiki/Charles_Hodgehttp://scratchpad.wikia...

>
> > I mention Behe only because it's his definition Behe and Miller use
> > when discussing Behe's IC mousetrap. �It's disingenuous for you to
> > disagree with Miller by substituting your own definition.
>
> > More to the point, "optimal" is a variable that depends on what's
> > being measured. �For example optimizing for thermal efficiency, speed,
> > and yield may require three entirely different designs. �Adding
> > "optimal" to IC adds more smoke than light.
>
> But adding "optimal" in as another constraint is handy when you aiming to
> find The Designer. �The more characteristics of design you can get
> incorporated into the analogy the more chance it will magically demonstrate
> there is a designer.
>
> >>> Behe says the separate parts must evolve simultaneously, and
> >>> that is something evolution can't do.
>
> >> Evolution in the pattern or design sense?
>
> > In whatever sense Behe and Miller used.
>
> This game over senses of some key words is the whole reason for starting
> this and every thread. �Don't go in there unless you want to see all your
> words sucked inexorably into the black hole of meaning at the centre of
> backspace's mind.
>
> David

Warning heeded.

Christopher Denney

unread,
Jan 29, 2011, 5:23:03 PM1/29/11
to
On Jan 29, 3:57 pm, jillery <69jpi...@gmail.com> wrote:
> On Jan 29, 4:39 pm, backspace <stephan...@gmail.com> wrote:
>
>
>
>
>
>
>
>
>
> > On Jan 29, 11:18 pm, Bob Casanova <nos...@buzz.off> wrote:
>
> > > On Sat, 29 Jan 2011 10:46:09 -0800 (PST), the following
> > > appeared in talk.origins, posted by backspace
> > > <stephan...@gmail.com>:
>
> > > >On Jan 29, 5:14 pm, jillery <69jpi...@gmail.com> wrote:
> > > >> 3. Since NS has no foresight, and can only select variations using
> > > >> present conditions, therefore NS, and so evolution, can't explain IC
> > > >> systems.
>
> > > >When you say NS has no ''foresight'' , do you mean the concept NS1 or
> > > >NS4?
>
> > > He means that selection acts without intent, on current
> > > populations in current environments. As you know very well.
>
> > One can enact a selection for cafe over banana. But to say
> > "....selection acts without intent....." is grammatically corerct but
> > as meaningless as Colorless green ideas sleep furiously. The entire
> > origins debate is bogged down with meaningless sentences.
>
> > 1) Athttp://bit.ly/gVwUAQtheposter wrote "...To evolve IC systems,

Oh he's been doing that a couple of times a year for as long as I can
recall. (for his sentences anyway)

Frank J

unread,
Jan 29, 2011, 5:21:06 PM1/29/11
to

Basically, he is reducing it to his level.

>
>
>
> > A most unfortunate event transpired in that a mother and wife was shot
> > by what seems to be a paranoid schizophrenic. With cunning and callous
> > precision this was used to target political opponents for ratings. A
> > schizophrenic can be influenced by over the top metaphorical imagery
>

> ...
>
> read more �- Hide quoted text -
>
> - Show quoted text -- Hide quoted text -
>
> - Show quoted text -


Robert Camp

unread,
Jan 29, 2011, 5:38:38 PM1/29/11
to
On Jan 29, 2:21 pm, Frank J <f...@verizon.net> wrote:
> On Jan 28, 9:01 pm, Robert Camp <robertlc...@hotmail.com> wrote:
>
> > On Jan 28, 3:28 pm, backspace <stephan...@gmail.com> wrote:
>
> > > On Jan 28, 8:49 pm, hersheyh <hershe...@yahoo.com> wrote:
>
> > > > On Jan 28, 10:51 am, John Harshman <jharsh...@pacbell.net> wrote:
>
> > > > > backspace wrote:
> > > > > > On Jan 28, 4:02 am, John Harshman <jharsh...@pacbell.net> wrote:
> > > > > >> Burkhard wrote:
> > > > > >>> On 27/01/2011 20:11, John Harshman wrote:
> > > > > >>>> backspace wrote:

<snip>

> > > Our political and social discourse have embalmed into hyperbole. From
> > > Glen Beck's acidifying oratory to the cornification of America by
> > > Sarah Palin and emotional slop from Oprah Winfrey the American mind
> > > has entered a type of intimate cartoonish universe, its sense of the
> > > spiritual and eternal oxidized.
>
> > "Acidifying," "cornification," "oxidized.?" I like a good turn of
> > phrase as much as the next guy, but do you actually have any idea what
> > you're trying to say?
>
> Basically, he is reducing it to his level.

Geez, you don't have to be sarcaustic about it.


backspace

unread,
Jan 29, 2011, 6:17:01 PM1/29/11
to

See http://en.wikipedia.org/wiki/Pragmatics which describes why no
word means anything, only ideas have meaning.

R.D. Heilman

unread,
Jan 29, 2011, 7:48:47 PM1/29/11
to
On 01/27/2011 05:19 PM, Mike Painter wrote:
> backspace wrote:
>> http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>>
>> === Irreducible optimized functionality ===
>> The achieve the function or purpose of catching mouse needs five
>> parts. There can't be less than five parts or the ''optimum''
>> functionality won't be achieved.
>
> Fiddle. I can build a mouse trap with four parts (three if you don't count
> the bait and two if placed in a proper location) that is far better than
> your five part model.
> The first one I built caught 30 mice in one night and required no attention.
>
That's nothing; with only two moving parts, I've caught rabbits,
a squirrel, two raccoons and one hawk. Not at the same time.

>
> Lessor live trap models have only two parts plus bait.
>
> Not only is your model false but it ignores a huge variety of mouse traps
> that have few parts and are more effective.
>
>
> I do wonder how the number five came up?
> 1. Spring. (all that is needed in a good design.)
> 2. bait holder/trigger release.
> 3. trigger
> 4 trigger fastener
> 5 bait holder fastener
> 6. spring holder
> 7. base platform.
> 8. bait
>

jillery

unread,
Jan 29, 2011, 8:00:22 PM1/29/11
to
> Seehttp://en.wikipedia.org/wiki/Pragmaticswhich describes why no

> word means anything, only ideas have meaning.

Thanks for sharing.

David Hare-Scott

unread,
Jan 30, 2011, 1:22:11 AM1/30/11
to
R.D. Heilman" <"R.D. Heilman wrote:
> On 01/27/2011 05:19 PM, Mike Painter wrote:
>> backspace wrote:
>>> http://scratchpad.wikia.com/wiki/Irreducible_Functionality
>>>
>>> === Irreducible optimized functionality ===
>>> The achieve the function or purpose of catching mouse needs five
>>> parts. There can't be less than five parts or the ''optimum''
>>> functionality won't be achieved.
>>
>> Fiddle. I can build a mouse trap with four parts (three if you don't
>> count the bait and two if placed in a proper location) that is far
>> better than your five part model.
>> The first one I built caught 30 mice in one night and required no
>> attention.
> That's nothing; with only two moving parts, I've caught rabbits,
> a squirrel, two raccoons and one hawk. Not at the same time.
>

You can catch mice with a trap that has no moving parts. It's called a
bucket bottle trap. Unlike mechanical traps it doesn't need to be re-loaded
after every catch.

D

backspace

unread,
Jan 30, 2011, 2:42:18 AM1/30/11
to

But I bet you won't be albe to mass produce and sell it .

jillery

unread,
Jan 30, 2011, 5:47:42 AM1/30/11
to

When I was young and dumb, my family caught geese with a properly-
sized hole in the ground and some corn tossed on the bottom. No
moving parts at all, unless you count the goose. Before it was
caught, both the goose and my family were hungry. Afterward, neither
the goose nor my family was hungry. AFAIK the geese we caught are
still non-hungry, so I could say we helped to reduce the amount of
suffering in the Universe.

Bob Casanova

unread,
Jan 30, 2011, 12:13:01 PM1/30/11
to
On Sat, 29 Jan 2011 13:39:40 -0800 (PST), the following

appeared in talk.origins, posted by backspace
<steph...@gmail.com>:

>On Jan 29, 11:18 pm, Bob Casanova <nos...@buzz.off> wrote:
>> On Sat, 29 Jan 2011 10:46:09 -0800 (PST), the following
>> appeared in talk.origins, posted by backspace
>> <stephan...@gmail.com>:
>>
>> >On Jan 29, 5:14 pm, jillery <69jpi...@gmail.com> wrote:
>> >> 3. Since NS has no foresight, and can only select variations using
>> >> present conditions, therefore NS, and so evolution, can't explain IC
>> >> systems.
>>
>> >When you say NS has no ''foresight'' , do you mean the concept NS1 or
>> >NS4?
>>
>> He means that selection acts without intent, on current
>> populations in current environments. As you know very well.
>
>One can enact a selection for cafe over banana. But to say
>"....selection acts without intent....." is grammatically corerct but
>as meaningless as Colorless green ideas sleep furiously. The entire
>origins debate is bogged down with meaningless sentences.

Have you ever considered that when a large number of people
understand something which you find to be meaningless the
fault may not be with the others?

<snip more word salad>

Bob Casanova

unread,
Jan 30, 2011, 12:15:45 PM1/30/11
to
On Sat, 29 Jan 2011 15:17:01 -0800 (PST), the following

appeared in talk.origins, posted by backspace
<steph...@gmail.com>:

>On Jan 30, 12:50 am, jillery <69jpi...@gmail.com> wrote:

Nice to see you agree that your words are meaningless. Be
sure to come back when you can express your incorrect ideas
without words.

backspace

unread,
Jan 30, 2011, 2:52:52 PM1/30/11
to
On Jan 30, 7:13 pm, Bob Casanova <nos...@buzz.off> wrote:
> On Sat, 29 Jan 2011 13:39:40 -0800 (PST), the following
> appeared in talk.origins, posted by backspace
> <stephan...@gmail.com>:

>
>
>
> >On Jan 29, 11:18 pm, Bob Casanova <nos...@buzz.off> wrote:
> >> On Sat, 29 Jan 2011 10:46:09 -0800 (PST), the following
> >> appeared in talk.origins, posted by backspace
> >> <stephan...@gmail.com>:
>
> >> >On Jan 29, 5:14 pm, jillery <69jpi...@gmail.com> wrote:
> >> >> 3. Since NS has no foresight, and can only select variations using
> >> >> present conditions, therefore NS, and so evolution, can't explain IC
> >> >> systems.
>
> >> >When you say NS has no ''foresight'' , do you mean the concept NS1 or
> >> >NS4?
>
> >> He means that selection acts without intent, on current
> >> populations in current environments. As you know very well.
>
> >One can enact a selection for cafe over banana. But to say
> >"....selection acts without intent....." is grammatically corerct but
> >as meaningless as Colorless green ideas sleep furiously. The entire
> >origins debate is bogged down with meaningless sentences.
>
> Have you ever considered that when a large number of people
> understand something which you find to be meaningless the
> fault may not be with the others?

Indeed I have considered this. Which is why I post my ideas and nearly
wrote an entire book consisting of 84 pages at
http://scratchpad.wikia.com/wiki/TauTology

If I am the only person agreeing with these views then that would be
very troubling. The whole point is to convince Ken Ham, Hovind,
Wilkins, Behe etc. that they were engaging in grammatically correct
but meaningless sentences. This has prevented us from fully exploring
their ideas on IC and IF for example.

But calling Ken Ham a greater danger to Christianity then Dawkins
isn't exactly going to endeer me to the Christian world. I don't say
such with any malice or ill-will but out of concern for Ken Ham's soul
and the Christians who's thinking he is scrambling in a blender.

David Hare-Scott

unread,
Jan 30, 2011, 4:12:14 PM1/30/11
to

Nah. No need to sell them, I rent them out to people who want excavation
done, specially roadworks.

David

deadrat

unread,
Jan 30, 2011, 5:18:59 PM1/30/11
to
On 1/29/11 8:05 AM, backspace wrote:
> On Jan 29, 11:27 am, BramH<br.hess...@planet.nl> wrote:
>>>> Behe says the separate parts must evolve simultaneously, and
>>>> that is something evolution can't do.
>
>> Consider gazelles being chased by cheetahs ( i get this example from
>> one of Ernst Mayr's books.)
>
>> Try to deny any of the following:
>
> Right here is your problem. In the immortal words of Darwin himself
> ''... the truth of the propositions cannot be disputed...'' Popper
> demonstrated that if your statements or theory can't be disputed(if
> there is no way to test them) they aren't theories but logical
> fallacies. A theory is defined as something which can be disputed. We
> are after a theory of derivation(Prof. Owen's Doctrine of Derivation)
> not a series of truisms from which follows a non-sequitur.
> We are told that due to Uranium decay the earth is 5bil years old.
> Then the Uranium is dated to 5bil years due to the earth being 5bil
> years old. This is circular reasoning, the earth is dated to 5bil
> using UR and then UR dated to 5bil using the earth. What we need is an
> independant means of dating the UR itself. Because we don't know how
> UR came into existance.

You're told that Uranium decay dates the earth because of the
well-established rules of exponential decay of Uranium atoms. These
rules have nothing intrinsic to them about the age of the earth.
<snip/>

deadrat

unread,
Jan 30, 2011, 5:32:16 PM1/30/11
to
On 1/29/11 3:39 PM, backspace wrote:
<snip/>

>
> Lets rephrase this in terms of PD or NS:
> 3) "...To evolve IC systems, the Gaia nature selection force enacts a
> Preferential Decision(Natural Selection) for the variant based on what
> that part will do in the future, after the IC system is assembled by
> Lord Gaia ....."
>
> Most users of the object NS won't intend 3)

Of course not. Because it's nonsense. The metaphorical term "Gaia
nature selection force" adds nothing as it's, well, metaphorical. Since
there's no real "force," the word "enacts" is misleading as there's no
entity that takes action. Natural selection is what happens to the next
generation in populations with differing members as these members
confront their environment. Some survive and some don't based on what
happens to them as they live out their lives. The survival of an
individual doesn't depend on future variations in the population.
<snip/>


deadrat

unread,
Jan 30, 2011, 5:35:05 PM1/30/11
to

Probably won't endear you to anyone either.

But why is it that soi disant Christians can say and do the cruelest
things to others while they maintain that their cruelty is without animus?

backspace

unread,
Jan 31, 2011, 12:40:15 AM1/31/11
to

How do you know the rate of decay was the same in the past .
http://scratchpad.wikia.com/wiki/Gravity_Past

backspace

unread,
Jan 31, 2011, 12:51:13 AM1/31/11
to
On Jan 31, 12:32 am, deadrat <a...@b.com> wrote:
> On 1/29/11 3:39 PM, backspace wrote:
> <snip/>
>
>
>
> > Lets rephrase this in terms of PD or NS:
> > 3) "...To evolve IC systems, the Gaia nature selection force enacts a
> > Preferential Decision(Natural Selection) for the variant based on what
> > that part will do in the future, after the IC system is assembled by
> > Lord Gaia ....."
>
> > Most users of the object NS won't intend 3)

> Of course not.  Because it's nonsense.  The metaphorical term "Gaia
> nature selection force" adds nothing as it's, well, metaphorical.  Since
> there's no real "force," the word "enacts" is misleading as there's no
> entity that takes action.

I rest my case....

>  Natural selection is what happens to the next generation in populations with differing members as these members
> confront their environment.

Which is a metaphorical way of saying 'what happens, happens' in your
materialist world view. In another world view it could be literally
meant as a Gaia Thor like consciouis nature force. Metaphors allows
for much ambiguity. The materialist world view is expressed in 95% of
cases using NS as a metaphor in an attempt to obfuscate that their
entire train of thought reduces to three words: what happens,
happnes. Here we are something had to happen.

With a meaningless sentence the dominant culture world view is
expressed, no exact meaning can be pinned to a meaningless sentence.
Like for example the word 'Darwinism' where Michael Ruse can directly
contradict John Wilkins and other authors using the same word.

Ruse said ''.... Darwinims is the mechanism....'' Others would say
this is wrong. They don't understand ''Darwinism'' is neither right
nor wrong but undefined.

> Some survive and some don't based on what happens to them as they live out their lives.  

Truism.

jillery

unread,
Jan 31, 2011, 12:57:09 AM1/31/11
to

I see a problem here, that you use words and sentences to argue that
words and sentences have no meaning. I admit words and sentences are
all you have to describe your POV, but then why criticize that same
use by others? ISTM the only distinction of any consequence here is
between words and sentences that correspond to specific ideas, and
words and sentences that don't, ie nonsense. In this specific case,
the phrase (not sentence) "natural selection" corresponds to an idea,
specifically distinct but analogous to artificial selection. You may
disagree with that idea on its merits, but to argue using words and
sentences that "natural selection" has no meaning is hypocritical and
specious.

> If I am the only person agreeing with these views then that would be
> very troubling. The whole point is to convince Ken Ham, Hovind,
> Wilkins, Behe etc. that they were engaging in grammatically correct
> but meaningless sentences. This has prevented us from fully exploring
> their ideas on IC and IF for example.
>
> But calling Ken Ham a greater danger to Christianity then Dawkins
> isn't exactly going to endeer me to the Christian world. I don't say
> such with any malice or ill-will but out of concern for Ken Ham's soul

> and the Christians who's thinking he is scrambling in a blender.- Hide quoted text -