# What has Gpuccio’s challenge shown?

(Sorry this is so long – I am in a hurry)

Gpuccio challenged myself and others to come up with examples of dFSCI which were not designed. Not surprisingly the result was that I thought I had produced examples and he thought I hadn’t.  At the risk of seeming obsessed with dFSCI I want assess what I (and hopefully others) learned from this exercise.

Lesson 1) dFSCI is not precisely defined.

This is for several reasons. Gpuccio defines dFSCI as:

“Any material object whose arrangement is such that a string of digital values can be read in it according to some code, and for which string of values a conscious observer can objectively define a function, objectively specifying a method to evaluate its presence or absence in any digital string of information, is said to be functionally specified (for that explaicit function).

The complexity (in bits) of the target space (the set of digital strings of the same or similar length that can effectively convey that function according to the definition), divided by the complexity in bits of the search space (the total nuber of strings of that length) is said to be the functional complexity of that string for that function.

Any string that exhibits functional complexity higher than some conventional threshold, that can be defined according to the system we are considering (500 bits is an UPB; 150 bits is, IMO, a reliable Biological Probability Bound, for reasons that I have discussed) is said to exhibit dFSCI. It is required also that no deterministic explanation for that string is known.”

(In some other definitions Gpuccio has also included the condition that the string should not be compressible)

These ambiguities emerged:

Some functions are not acceptable but it is not clear which ones.  In particular I believe that functions have to be prespecified(although Gpuccio would dispute this). Also functions which consist of identifying the content of  “data strings” (a term which is itself not so clear) are not acceptable because the string in question could have been created by copying the data string.

The phrase “no deterministic explanation for that string is known” is vague.  It is not clear in how much detail and how certainly the deterministic processes have to be known. For example, it appears from above the possibility that the string in question might have been copied from the string defining the function by some unknown method is sufficient to  count as a known deterministic explanation. This implies that really it is sufficient to be able to conceive of the very vague outlines of a determinist process to remove dFSCI. I think this amounts to another implicit condition: no causal relationship between the function and the string.

Lesson 2)  dFSCI is not a property of the string.

It is a relationship between a string, a function and an observer’s knowledge. Therefore, it may be that dFSCI applies for a string for one observer with a certain function but not for another observer with a different function.  The rules for deciding which function are not clear.

Lesson 3) The process for establishing the relationship 100% specificity of dFSCI and design is not commonly found outside examples created by people to test the process.

“To assess the dFSCI procedure I have to “imagine” absolutely nothing. I have to assess dFSCI without knowing the origin, and then checking my assessment with the known origin.”

When challenged he was unable to name any instances of this happening outside the context of people creating or selecting strings to test the process as in our discussions. This is important as the dFSCI/design relationship is meant to be an empirical observation about the real world applicable to a broad range of circumstances (so that it can reasonably be extended to life). If it is only observed in the very special circumstances of people making up examples over the internet then the extension to life is not justifiable. To give a medical analogy. It might well be that a blood test for cancer gives 100% specificity for rats in laboratory conditions. This is not sufficient to have any faith in it working for rats in the wild, much less people in the wild. Below I discuss what is special about the examples created by people to test about the process.

A Suggested Simplification for dFSCI

dFSCI says that given an observer and a digital string where:

1) The observer can identify a function for that string

2) The string is complex in the sense that if you just created strings “at random” the chances of it performing the function are negligible

3) The string is not compressible

4) The observer knows of no known deterministic explanation for producing the string

Then in all such cases if the origin eventually becomes known it turns out to include design.

Given the rather lax conditions for “knowing of a deterministic mechanism” that emerged above, surely  (2) and (3) are  just special cases of (4). If (2) or (3) were present then deterministic mechanisms would be conceivable for creating strings.

So the dFSCI argument could be restated:

Given an observer and a digital string where:

* The observer can identify a function for that string

* The observer cannot conceive of a deterministic explanation for producing the string

Then in all such cases if the origin eventually becomes known it turns out to include design.

Conclusion

There are two main objections to the ID argument:

A) There are deterministic explanations for life.

B) Even if there were no deterministic explanations it would not follow that life was designed

For the purposes of this discussion I will pretend (A) is false and focus on (B)

No one disputes that it is possible to detect design.  The objectors to ID just believe that B) true. The correct way of detecting design is to compare a specific design hypothesis with alternatives and assess which is provides the best explanation. This includes assessing the possibility of the designer existing and having the motivation and ability to implement the design.   If no specific hypothesis is available then nothing can be inferred.

So is the dFSCI claim above true and if so does it provide a valid alternative way of detecting design?

The trouble is that there is dearth of such situations. One of the reasons for this is that digital strings do not exist in nature above the molecular level.  At any other level it is only a human interpretation that imposes a digital structure on analogue phenomena.  The characters you are reading on this screen are analogue marks on the screen. It is you that is categorising them into characters. So all such strings are created by human processes. It follows that design is a very plausible explanation for any such string.  People were involved in the creation and could easily have designed the string. If you add the conditions that the function must be prespecified and there should be no causal relationship between the function and the string then design is going to be by far the best explanation. It goes further than that.  It also means there almost no real situations where someone is confronted with a digital string without knowing quite a bit about its origin – which is presumably why Gpuccio can only point to examples created/selected by bloggers.

What about the molecular level?  Here there are digital strings that are not the result of human interpretation. Now human design is massively implausible (except for a few very exceptional cases).  The problem now is that carbon chains are the only digital strings with any kind of complexity and these are just the one’s we are trying to evaluate. There are no digital strings at the molecular level with dFSCI except for those involved in life.

So actually the dFSCI argument only applies to a very limited set of circumstances where a Bayesian inference would come to the same conclusion.

## 493 thoughts on “What has Gpuccio’s challenge shown?”

1. When ID types come up with arguments that are, or seem to be, quantitative, analyses of these arguments need to appear here (or at other relevant forums such as Panda’s Thumb). I’m going to keep doing that.

Of course I am not assuming that the hard-core ID posters and commenters will be convinced. They never acknowledge that any of their arguments are wrong.

But there are numerous spectators out there reading these threads, and they need to hear how theoretical and quantitative arguments work, and I can supply that.

There are also other TSZ and PT commenters who don’t know how to evaluate these putatively-quantitative ID arguments. These commenters seem to find it useful to have a careful explanation of the theory.

We need an accessible place for these quantitative and theoretical arguments to occur. TSZ and PT are such places.

2. gpuccio: “e) If you succeed, you will have “invalidated “dFSCI” as a design detection tool”.

OK?”

gpuccio, you have to start posting comments on this side if you want a reply.

It’s not fair to readers to have to follow two blogs for one conversation.

Is your “dFSCI” important enough to be defended in a cold objective non-biased environment?

3. Gpuccio,

I have given you an explicit example that shows why the design hypothesis doesn’t predict an objective nested hierarchy. In your response, you carefully avoid my example and merely reassert the claim that I had just debunked:

“The only “assumption” necessary to explain that kind of nesting in the design explanation is that the designer needs to act through common descent, and to reuse what already exists with intelligent modifications. It seems not such an extreme assumption, and it fits the facts.”

First of all, you haven’t given any independent justification for your assumption. A designer (and especially a Designer) doesn’t have to work through common descent, and he doesn’t have to reuse what already exists. Your only reason for assuming that he does these things is that you are trying to force-fit your theory to the existing evidence. It’s the same error made by an advocate for the Rain Fairy hypothesis who assumes that the Rain Fairy always acts in ways that match the weather we are actually observing.

Second, your assumption doesn’t even work. The example I gave, in which a computer designer takes a Blu-ray drive (which already exists) and adds it to 25 different computer models (which already exist), thus intelligently modifying them, conforms to your assumption. Yet I have already shown that it spoils the objective nested hierarchy.

Third, the Blu-ray example is easily translated into a biological context. The Designer takes a complex structure that he implemented elsewhere (analogous to the Blu-ray drive) and “transplants” it into 25 distinct lineages (analogous to the 25 separate computer models). It’s design, it conforms to your unjustified assumption, and yet it still ruins the objective nested hierarchy.

You have a serious problem. The theory of evolution makes the audacious prediction of an objective nested hierarchy, out of trillions of alternative possibilities. The prediction is confirmed. Under a design hypothesis, you have no reason to expect an objective nested hierarchy.

If you “follow the evidence where it leads”, as IDers like to say, you find that it leads directly away from design and straight to modern evolutionary theory.

4. You appear to be making a Bayesian argument, and Bayesian arguments are  off limits according to Dembski, unless it is convenient for him (as in his analysis of election fraud).

5. Joe F.,

Gpuccio confirms that he is still confused about the difference between the model and the thing being modeled:

My idea was:

a) we take some computer environment with natural computer resources, designed in blind by people who are not aware of the implementation we are going to use it for.

b) we design replicators that can well reproduce in that environment (like computer viruses) using those natural resources.

c) we introduce in the replicators some random variation mechanism, which can be modulated appropriately to test different rates and modalities of random variation.

d) we just wait.

e) if and when the replicators develop new functions, we check their code to see if the new function is complex.

That would be an implementation of NS in a computer environment, and could be considered in some way an indicator of what NS can or cannot do, even if obviously a computer environment remains different from a biological environment.

What he’s suggesting is not a model of NS, it’s an “implementation of NS” — his words. He’s rejecting the idea of modeling selection and saying instead that we must actually implement selection by creating computer viruses which compete for computational resources, which he oddly labels “natural resources”.

The obvious irony (obvious to us, anyway) is that in trying to model NS, he ends up implementing selection, rather than modeling it. And it’s a form of selection that isn’t derived from actual natural fitness landscapes. Instead, it’s based on the success of artificial replicators in an artificial environment with an artificial source of variation.

He seems to understand that it is perfectly fine for the programmer to model variation:

c) we introduce in the replicators some random variation mechanism, which can be modulated appropriately to test different rates and modalities of random variation.

Yet for some reason he doesn’t understand that if it’s okay to design a model of random variation, then it’s okay to design a model of natural selection.

Anyway, once you cut through all of his confusion, the bottom line is this: Random variation plus selection can produce functional complexity, as gpuccio himself admits. Whether it does so in a particular case depends on the specific fitness landscape involved.

Gpuccio is therefore betting the farm on the hope that actual biological fitness landscapes will consist of separated “islands of function” which NS cannot navigate.

The evidence for the objective nested hierarchy demonstrates that this is a huge mistake. Studies by Szostak, Lenski, Thornton, etc., are additional nails in the coffin.

6. http://toddcwood.blogspot.co.uk/2012/10/biochemistry-tour-de-force-in-nature.html

Their results confirm some of what protein biochemists have long believed about protein sequence and function.  Substitutions at most sites (about 75%) had little effect on the protein function, and some sites tolerated substitutions of very dissimilar amino acids with little to no functional consequence.  More surprisingly, these weren’t always surface amino acids.  At least one site that has direct contact with the binding partner was quite tolerant of substitutions.  Only 20 amino acid sites were sensitive to substitutions, and that’s about a fourth of the amino acid positions in the protein.

7. Forums such as Panda’s Thumb and The Skeptical Zone can be good places to learn science as well as for debunking ID/creationism.

Debating ID/creationists may get one some insight into their tactics; but those tactics are always variations on the old “Gish Gallop;” they will never converge because their only purpose is to make it appear that the ID/creationist can stay in the game and take on multiple opponents at a time.  If they can do that – no matter how stupid their on-the-spot concocted “arguments” – they win.

I suppose it is because I have been observing this for as long as I have that so much of it seems like a time loop to me.  I seldom see anything new in ID/creationist debating tactics.  It’s always the same smoke-and-mirror word-gaming; never any real substance or any real science.

The focus is always on making the ID/creationist a celebrity on the spot.  ID/creationists crave celebrity; and Ken Ham is falling all over himself recently in order to get Bill Nye to debate someone on the AiG staff.  This has always been what they want; they simply cannot do science.  I have seen up close and personal just how damned dishonest and mean that ID/creationists can be.  Early on, Gish was one of the worst; but other creationists copied him and his tactics.  Those tactics were taught explicitly at places like the ICR and at Liberty “University.”  They are not nice people despite their façade.  They are down and dirty politics to the core.  Debating is their specialty.

What I learned years ago was that the public – kids especially – really appreciate honest attempts of scientists to clarify science in a way that doesn’t condescend and yet makes the science interesting and useful in their lives.  They also want to understand the mechanics behind deception; and the story of ID/creationism is a perfect case study in conscious, well-organized and well-coordinated deception.  The lessons of scientific thinking are also lessons in minimizing the dangers of being deceived.

ID/creationists don’t need to know who the effective educators are; they don’t need to know who to send their assassins after.  ID/creationists are always looking for the high-profile, celebrity scientist on which to hitch a ride.  They need to be taken down by nobodies coming out of nowhere. Good public education is necessary.

The last thing an ID/creationist should ever see when he is trying to peddle his crap is the cold, knowing glare of the public slamming the door in his face with no comment.  He should never know where that knowledge came from.

8. Even using gpuccio’s unrealistic and illogical limitations, it seems that Tom Ray’s tierra system meets his requirements.  This was raised both at Mark Frank’s blog where we got to watch him change definitions on an almost hourly basis and in an earlier thread here at TSZ when gpuccio deigned to grace us with his presence.

He seems oddly disinclined to discuss that particular simulation.  Too bad, it would be nice to see an intelligent design creationist make a testable prediction for once.

9. Mike Elzinga:

Debating ID/creationists may get one some insight into their tactics; but those tactics are always variations on the old “Gish Gallop;”

That’s simply not true. Gpuccio, for example, doesn’t use the Gish Gallop. (He changes his argument frequently, often without acknowledging that he is doing so, but the topic remains the same, and so this is not Gish Galloping.)

Also, I’m not interested merely in tactics. The thinking of IDers and creationists is also fascinating. Many of them can see that reason and evidence are at odds with their beliefs. How they manage the dissonance, and how they maintain their beliefs in the face of contradictory evidence, is very interesting.

10. By “ID/creationists” I am assuming you mean debaters in public, including on blogs. Some such qualification is in order, otherwise you are accusing between 25 and 50% of the U.S. population of dealing dishonestly with these issues. The great bulk of those people are of course sincere and honest in their beliefs about creationism.

11. Shutdown TSZ – no.

But could we have people starting new topics – please? They don’t have to be based on UD posts.

12. Also, with gpuccio it is sometimes possible to zero in on the crux of a disagreement. You can’t do that with Gish Gallopers.

For example:

keiths:

Gpuccio is therefore betting the farm on the hope that actual biological fitness landscapes will consist of separated “islands of function” which NS cannot navigate.

gpuccio:

Yes.

13. Almost certainly real biological fitness landscapes do have well-separated islands. Probably all life exists on one of those islands, the others being unknown to us. There might be much better ways to be life, but life never got there.

The issue is not whether there are such islands, but there is more than one such island with present-day life forms on it.

14. Yup; primarily the dispensers and political pushers of this crap.

I think I may have mentioned over on Panda’s Thumb that Duane Gish used to show up unannounced in the biology classrooms in the public schools around Kalamazoo, MI and harass the biology teachers.  This was way back when Gish worked at what used to be the Upjohn Company (now Pharmacia).   With the changes in the law since then, he could never do that today.

One of those teachers was a multiple award winning teacher who was a good friend of mine for many years.  The pain of those encounters with Gish lasted for many years, and she still recounted them when I last talked with her a few weeks before she died of cancer several years ago.

Think of it.  What kind of man would attempt to politically assassinate a biology teacher right in front of his or her students?  But this is the kind of man Gish was; and he had support in the Kalamazoo community. The churches that supported him are still here; and they invite Ken Ham and his minions from time to time.

Some of the local biology teachers, even today, are skittish about teaching anything about evolution; and even the biology teachers at a highly competitive math/science center get harassed regularly about their teaching of evolution.  One of the community’s former members of the Michigan State House of Representatives is still politically active and pushed ID/creationist legislation when he was in the House.  He is a Tea Party member today.

During the Dover trial, one of the nearby communities was threatened with a law suit by the Thomas Moore Law Center if it didn’t allow two elementary public school teachers to continue teaching young earth creationism to the students.  That battle lasted for nearly a year.

This is not the Deep South.  These people are out there; and they are still angry and active.  I have responded to their diatribes in the local newspaper as recently as 2008 or 9.  They were expecting much from this last Presidential Election; but fortunately they lost.  But they never quit.

When you meet some of them, they can be very cordial; but they are politically nasty behind the scenes.  They try not to tip their hands much any more.  I’m glad the election turned out the way it did; but it’s not over.

15. Even if we use `tierra`, there is still this issue:

[gpuccio #888] e) if and when the replicators develop new functions, we check their code to see if the new function is complex.

We would need to agree on what was or was not a “new function”. And it would not suffice to examine the code that made the new function immediately after it arose, as if it become “complex” later, by elaboration, that would seem to satisfy the requirements of dFCSI.

16. Joe F:

Probably all life exists on one of those islands, the others being unknown to us. There might be much better ways to be life, but life never got there.

The issue is not whether there are such islands, but there is more than one such island with present-day life forms on it.

And of course, gpuccio thinks that life is spread over multiple islands, and that it had to be poofed onto each one since unguided evolution is incapable of bridging the gaps between them.

17. Certainly I encourage folks to study ID/creationist tactics.  Debating is not as efficient as sitting down and really dissecting their misconceptions, misrepresentations, and tactics.

But ID/creationists who want to debate aren’t really interested in science; and from what I have seen for many years, I doubt that there are very many who can follow any real science.  Most of them have vague, hand-waving ideas they pick up from creationist literature.

What I have seen of them is that they want the practice so that they can win debates.  Many declare victory even when they loose.  They never learn any science from a debate.  By just getting a debate, they think they win.

However, as long as there are people who plunge into debating them, if I happen to see anything interesting, I use the occasion to simply sit back and study.  I am generally not interested in debating ID/creationists because their understanding of science is not adequate enough for them to understand any point one is making.  They don’t want to learn anyway.  It’s just the debate for them.

I personally prefer helping people to understand science; and I don’t like the game-playing of debates.  Debates are horribly inefficient at conveying reliable information.  ID/creationist debaters are a waste of time for me.

If one is to engage in debates, I would certainly suggest that one come away from such encounters with the ability to do a thorough post-mortem of the debate as far as content, concepts, and one’s own understanding of science.

For the scientist, it isn’t enough just to win a debate; one has the additional responsibility of educating the audience as well.  Scientists are at a disadvantage in this regard because they are expected to be responsible for not misleading people.

ID/creationists don’t care; they can make up crap as they go and be glib enough to bamboozle the audience into thinking they are great.  It is never a fair match from the beginning; and they already know that.

Debates on the internet are a little different because there is a “transcript” generated; but when threads get as long and and as drawn out as they do here, for example, they stop holding interest, and nothing is every learned by the ID/creationist or any of the spectators who get bored and walk away.

I quite franky got bored with all this gpuccio stuff.  Learning something usually doesn’t take that long in real life. When you sum it all up, what have you learned?  What else could you have learned with all that time?

18. These days I usually ignore Mung, except when he a) displays a misunderstanding that I think others might share, or b) says something that is too funny not to point out. This is an example of the latter:

keiths:

First of all, you haven’t given any independent justification for your assumption. A designer (and especially a Designer) doesn’t have to work through common descent, and he doesn’t have to reuse what already exists.

So let’s label the assumptions and see who is making them:

keiths_assumption_01: A designer (and especially a Designer) doesn’t have to work through common descent.

keiths_assumption_02: he doesn’t have to reuse what already exists.

Assumption Score: gpuccio: 0 keiths: 2

Let’s take a closer look at Mung’s logic. Suppose gpuccio claims that the designer 1) works through common descent, 2) reuses what already exists, 3) makes changes only on the third Tuesday of each month, 4) fiddles with eukaryotes in the spring and summer and prokaryotes in the fall and winter, and 5) appears to earthly observers in the form of a giant, diamond-encrusted lobster.

I object, saying that the designer doesn’t have to work through common descent, doesn’t have to reuse what exists, doesn’t need to restrict changes to the third Tuesday of each month, can fiddle with both eukaryotes and prokaryotes year round, and doesn’t have to present the appearance of a diamond-encrusted lobster.

By Mung’s logic, I have made five assumptions and gpuccio has made none. Too funny.

19. And here are a few Mungish misconceptions that are worth rebutting because others might share some of them:

keiths:

Third, the Blu-ray example is easily translated into a biological context. The Designer takes a complex structure that he implemented elsewhere (analogous to the Blu-ray drive) and “transplants” it into 25 distinct lineages (analogous to the 25 separate computer models). It’s design, it conforms to your unjustified assumption, and yet it still ruins the objective nested hierarchy.

Yet another ‘test’ of evolutionary theory that isn’t.

If a designer did that, you would claim that the lineages were not distinct because they all shared this feature in common.

No, because lineages can share features and still be distinct. Think of different eukaryote lineages, for example.

And what your describing is pretty much what happens with HGT isn’t it?

No. We don’t see complicated structures being “transplanted” simultaneously and identically into 25 distinct lineages. That would be much too improbable to happen by chance.

But you don’t attribute that [HGT] to design, do you.

No, because HGT is explained by known unintelligent mechanisms. As gpuccio will tell you, it’s a mistake to infer design when something can easily be explained by non-design mechanisms.

And that invalidates your nested hierarchy, so you ignore it, as Joe has pointed out repeatedly.

No. I’ve stated repeatedly that HGT has to be limited in order for an objective nested hierarchy to be inferrable. Note that in Theobald’s example of the 30 taxa, prokaryotes are confined to a single taxon. Thus HGT among prokaryotes doesn’t obscure the objective nested hierarchy.

Are you saying convergence is proof of design? No, you wouldn’t say that.

If you found a case like the one I described, it would be evidence in favor of design. No one has found such a case, however.

Your theory is immune from disconfirmation.

Quite the opposite. Unguided evolution predicts the objective nested hierarchy. There are trillions of times more ways for this prediction to be falsified as there are for it to be confirmed. Evolution sticks its neck out and survives unscathed. ID can’t compete.

keiths:

He seems to understand that it is perfectly fine for the programmer to model variation:

No, that would be an implementation of variation, not a model of it.

Not according to gpuccio. See comment #913 at UD.

20. Intelligent selection [as opposed to natural selection] is any kind of selection where a conscious intelligent being decides what to select, and what the effects of selection will be.

A GA can be accepted, if it models well a real situation that can be observed. Obviously, the GA is designed in all cases, but we can restrict our analysis to what the GA does, not to its existence.

Gpuccio,

Those statements contradict each other. You need to retract one of them.

If statement #1 were true, it would mean that a fitness function could never legtitimately be included in a model of natural selection. That is obviously false, as you yourself have admitted.

I therefore recommend retracting statement #1.

21. You seem to know things about the designer that I don’t know.

“A designer (and especially a Designer) doesn’t have to work through common descent, ”

Why? This is really a bold assumption.

As I explained to Mung, you are the one making the assumptions. You state that “the designer needs to act through common descent”. How do you know that?

The answer, of course, is that you don’t. But you need to make that assumption, because otherwise your hypothesis doesn’t fit the evidence and gets tossed out.

A Rain Fairy proponent could make the same flawed argument that you are making:

Why do we see an objective nested hierarchy?

Because gpuccio says the Designer always acts (whether by choice or by limitation) in a way that produces an objective nested hierarchy.

How does gpuccio know this?

He doesn’t. But if he doesn’t make that assumption, then the design hypothesis doesn’t fit the evidence and gets tossed out in favor of unguided evolution, which fits the evidence exactly.

Why does rain happen in low pressure areas, and why do low pressure systems rotate in opposite directions in the northern vs. southern hemisphere?

Because this is how the Rain Fairy operates (whether by choice or by limitation), according to the Rain Fairy advocates.

How do the Rain Fairy advocates know this?

They don’t. But if they don’t make that assumption, then the Rain Fairy hypothesis doesn’t fit the evidence and gets tossed out in favor of modern meteorological theory, which fits the evidence exactly.

Gpuccio, isn’t it a little embarrassing to be making the same argument as the Rain Fairy advocates?

22. This could be summed up by saying that ID proposes an unobserved, discontinuous process to replace a regular process that is routinely observed, without proposing any agent capable of performing this activity,

All predicated on the truth of the isolated island metaphor,

It’s really that simple.

Now can we please have a thread discussing the island metaphor?  Everything else about ID is superfluous.

23. Petrushka:

This could be summed up by saying that ID proposes an unobserved, discontinuous process to replace a regular process that is routinely observed, without proposing any agent capable of performing this activity,

All predicated on the truth of the isolated island metaphor,

That’s pretty much it. I would only add a few things:

Gpuccio’s argument is an attempt to rescue ID.

ID comes into the fight with a huge disadvantage: it makes no predictions. Evolutionary theory, by contrast, makes risky predictions that are confirmed by the evidence. This includes the prediction of an objective nested hierarchy, which is validated to a spectacular degree of accuracy.

Gpuccio’s only hope in this lopsided battle is to show that unguided evolution doesn’t really explain the objective nested hierarchy. The “islands of function” assumption is intended to accomplish that. Unfortunately for gpuccio, he can’t justify the assumption, so he is left arguing that until the fitness landscapes are mapped in sufficient detail, the default must be to assume that life occupies isolated islands.

Even if he could justify the “islands of function” claim, gpuccio still has a problem: ID doesn’t fit the evidence. He makes additional unjustified assumptions to correct this.

The complete list of assumptions he must make:

1. Assume that life occupies isolated islands in the fitness landscapes, so that unguided evolution is ruled out.

2. Assume that a designer exists at the appropriate times and places.

3. Assume that the designer has the necessary capabilities to do the job.

4. Assume that the designer just happens to operate in a way that mimics unguided evolution (if unguided evolution were not ruled out by #1).

Gpuccio’s argument depends on all of these assumptions. He hasn’t justified any of them.

Funny you should say that. I’ve been working on one and hope to post it tonight or tomorrow.

24. You could certainly do worse than start with Darwin’s own formulation of the problem.

IF it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down. But I can find out no such case. No doubt many organs exist of which we do not know the transitional grades, more especially if we look to much-isolated species, round which, according to the theory, there has been much extinction. Or again, if we take an organ common to all the members of a class, for in this latter case the organ must have been originally formed at a remote period, since which all the many members of the class have been developed; and in order to discover the early transitional grades through which the organ has passed, we should have to look to very ancient ancestral forms, long since become extinct.

25. Keiths:

Also, with gpuccio it is sometimes possible to zero in on the crux of a disagreement. You can’t do that with Gish Gallopers.

For example:

keiths: Gpuccio is therefore betting the farm on the hope that actual biological fitness landscapes will consist of separated “islands of function” which NS cannot navigate.

gpuccio: Yes.

I suppose that is in a way a compliment. Thank you…

Re KS: Gish Gallop is of course a highly loaded and polarisingly denigratory personal attack, to explain away not having an answer. If you have evidence of improper argumentation give it, if not, stop accusing improperly. KF

Hey, kairosfocus, read my comment. I didn’t accuse gpuccio of being a Gish Galloper.  I defended him from the charge.

You, however, accused me of something I didn’t do. If you were in my position right now, you would be demanding an apology, as we all know from long experience. Since the tables are turned, will you be offering an apology, or will you choose instead to be a hypocrite?

26. I therefore recommend retracting statement #1.

gpuccio:

No, I think we can “zero in on the crux of this disagreement”. I maintain statement number 1.

So, what about a GA that models NS? It is a model which implements parameters appropriate for what is being modeled. It will use intelligent selection, but giving it a mathematical form which mimics true natural selection as we can observe it in nature. In that sense, it can give useful information.

Gpuccio,

You’ve changed your position yet again without acknowledging it. Reread this comment of yours and note the bolded portion:

To onlooker (at TSZ):

Any fitness function in any GA is intelligent selection, and in no way it models NS.

Please, do not consider any more that statement. Keiths is right, it was a wrong generalization. I have given a very generical example of how a fitness fuinction could be built that, while being essentially useless, at least would not be an implementation of IS, and could resemble generically what we expect from NS (IOWs, it would add nothing to what we already know).

The correct concept is as follows: It is completely wrong to model NS using IS, because they have different form and power.

27. To be able to explain that by a designer (the only credible explanation available) I have to assume that the designer worked in a way that explains the nested hierarchies.

And in particular, the objective nested hierarchy. The problem is that you know nothing about the designer, so you can’t independently justify your assumptions. He’s an unknown designer (maybe more than one), with unknown abilities, unknown limitations, and unknown goals.

The only “justification” you offer for your assumptions is that you want to fit your hypothesis to the evidence of the objective nested hierarchy. But then you’re committing the Rain Fairy fallacy.

It’s the equivalent of this:

You are arguing against a Rain Fairy proponent. You point out that under the Rain Fairy hypothesis, there is no reason to expect low-pressure systems to rotate in opposite directions in the two hemispheres. Any wind pattern is compatible with the Rain Fairy hypothesis, so it is completely arbitrary to assume that the Rain Fairy just happens to choose (or is somehow limited to choosing) the counter-rotating scheme.

Your opponent responds, “I am making a simple and reasonable assumption. I assume that the Rain Fairy likes symmetry, and this is why low-pressure systems rotate in opposite directions in the two hemispheres.”

Are you persuaded by the Rain Fairy advocate? If not, then why should we be persuaded by your argument, which is logically identical?

To summarize, the problem is that if you know nothing about the designer, you can’t independently justify any assumptions you make about him. And if you try to force-fit your hypothesis to the evidence by simply tacking on arbitrary assumptions, then you are committing the Rain Fairy fallacy.

28. Even darwinists use the concept of “convergent evolution”, which is in itself a violation of your rule. Maybe it simply describes cases where design was repeated in different ways, again violating what you say.

No, convergent evolution is not a violation of my rule. I wrote:

We don’t see complicated structures being “transplanted” simultaneously and identically into 25 distinct lineages. That would be much too improbable to happen by chance.

Complicated convergent structures are not genetically identical and they are unlikely to happen simultaneously. To get complicated, identical structures in 25 distinct lineages requires design.

29. You’re being too specific, I think. The Rain Fairy fallacy applies generally to gods, who are typically unconstrained and undefined, and can be used as a post facto explanation for anything. I defy you (or gpuccio or anyone at UD) to point to anything whatsoever, such that “the Designer chose to do it that way” wouldn’t be a “perfectly reasonable explanation”.

30. The Rain Fairy fallacy applies generally to gods, who are typically unconstrained and undefined, and can be used as a post facto explanation for anything.

Absolutely. That’s my point, but I find that people “get it” better when I use a specific example that they immediately recognize as ridiculous, such as the Rain Fairy.

Another one I like to use is the hypothesis that angels are pushing the planets around, but that they just happen to choose the same paths that gravity would have produced if it were operating in the solar system.

31. My observation is that “convergent evolution” is easily misrepresented. At a high enough level of abstraction, everything converges. Motility is common in organisms, so motility is convergent evolution. Where motility is not a strategy, then being stationary is convergent.

Get a bit more specific – look at flight, or at sight. Both are very useful, so one would expect numerous very different approaches to these abilities. And so on.

Where “convergent evolution” breaks down is where we go from the functionality level to the mechanical level. Multiple similar solutions to the same general problem is qualitatively different from multiple mechanically identical solutions.

I don’t think creationists want to understand the difference between homologies at the structural level, and similarities at the functional level. Birds, bats and bees all fly. HOW they fly is completely different. Hummingbird flight is MUCH more like eagle flight than like bumblebee flight.

32. Well, I’ve never seen anyone get through whether using general principles or specific examples. Personally, I think it’s reflected in the terminology. If you speak of garden fairies, or zeus, or the Great Green Arkleseizure, they know it’s silly fiction so they ignore it. If you attribute all the same characteristics but ascribe them to God, suddenly the toilet training kicks in, their god is REAL (when spelled right), and all manner of idiocy becomes “real”.

To those not infancy-trained to associate the word-icon “god” with indubitable certainty, with magic, with explicit support of whatever they might prefer to be true, everything the UD people argue is clearly insupporable. For those trained from infancy, they CAN not see it no matter how deeply your cram their face into it.

So I kind of feel sorry for gpuccio. I agree that he has recognized that his arguments fail, and the more he tries to reposition them, the more they fail. He is unhappy because he KNOWS his arguments MUST be true, god said so, but he just can’t find any way to explain it that isn’t arrant nonsense from the start.

The same pattern informs all of the UD arguments. They are all roundabout ways to find their non-negotiable god hidden SOMEWHERE in a perverse reality that needs no such hypothesis. I think I’d be depressed if I could find NO direct evidence for what I was trained to believe was true, and had to make up threadbare arguments and defend them with censorship, doubletalk and bafflegab.

33. It is very peculiar to read this kairosfocus character’s comments on the Gish Gallop.  He is a master at the Gish Gallop himself.

I was looking in on the discussion of entropy and thermodynamics a few of months ago when Sal Cordova posted his disagreement with Granville Sewell over on UD.  Of course, the KF character jumped right in with tons of copy/paste material.

This KF character’s responses were classic Gish Gallop.  KF dumped boatload upon boatload upon boatload of totally bogus material on several threads over at UD when all that “discussion” on entropy was going on.  Some of it was simply crap that he made up out of thin air, and other stuff was simply copied from sources he quote-mined without comprehension.

One seldom sees such a witch doctor caricature of mumbo-jumbo voodoo dumped in such a copious manner in order to overwhelm the discussion and give the appearance of great erudition and deep understanding.  And all of it was thrown in with an air of confident authority and condescension. The kindest thing I can say about KF is that he is a complete, double-down, triple-down fraud.

But that is the nature of UD; and that is why one should study them and not try to debate them.  They have no clue; but they use every Gish tactic developed at the ICR to bully and bluster as though they are experts in all things.  That has been the case ever since Morris and Gish started the ICR; and it has been ever downward and dirtier from there.  The tradition continues at UD.

Now those people over at UD are trying to rewrite their own history using the same tricks.  It isn’t going to happen.

34. keiths wrote (I leave out the details):

Gpuccio,

You’ve changed your position yet again without acknowledging it.

That is a pretty noticeable example of a self-contradiction. Wonder what gpuccio will say.

But in any case there is another contradiction. Some hundreds of messages ago, gpuccio and other commenters at UD were raising a completely different objection. Not to the fitnesses being unnatural. But that using a computer algorithm to model the reproduction of a population could not investigate whether CSI could be put into the genome by NS because the CSI is already there in the code that reproduces the digital organisms.

Now they seem to have a different objection, the one about naturalness of the fitnesses, and they do say that some kinds of simulation might be able to investigate this. Have they abandoned that earlier argument?

35. That is a pretty noticeable example of a self-contradiction. Wonder what gpuccio will say.

I doubt it’ll be “oh, I was wrong, the whole idea was flawed from the start, can we start over”?

IDists believe in design despite the lack of evidence, not because of it.

36. I don’t think simulations can address the actual fitness landscape or whether islands are isolated. That still needs to be done by Lenski, Thornton, et al.

What simulations can do is explore the abstract mathematical argument, and that argument is DOA.

What Axe, Behe, Dembski and that tribe have done is take the last series of Lotto winners and argue that there must have been fraud involved, because the odds against those particular people winning, taken as a group, are prohibitive.

In the world of DNA, they pretend that the path taken was the only possible path.

37. Did anyone else notice Cornelius Hunter’s blog where a commenter, RobertC, linked to Todd Wood’s take on this paper by McLaughlin et al? Seems yet more confirmation that fitness landscapes are not so rugged as some have surmised.

38. That Nature paper should come as no surprise; basic textbooks in organic chemistry have made this point for many decades.  Function and behavior are more often related to structure rather than the specific constituents.

And there is a good physical reason behind that; it has to do with the way that structure and charged distribution in complex systems of molecules are interrelated.

Also, increasing smoothness in the shapes of potential wells always occurs with increasing complexity.  Binding energies among constituents decrease with complexity.

The “smoothness” of potential wells is also related to the relative sizes of the potential energy well depths and the thermal kinetic energies of the bound constituents. If it is soft matter, the wells are smooth.

Fitness landscapes are directly related to the binding energies of the atoms and molecules that make up the organism.  It’s all basic physics and chemistry.

39. Did anyone else notice Cornelius Hunter’s blog where a commenter, RobertC, linked to Todd Wood’s take on this paper by McLaughlin et al?

This is the paper I hoped would form the basis of a new thread on islands of function. It looks like the creationists have responded in a creative way. Because most substitutions don’t have any effect, proteins can’t evolve.

40. … would form the basis of a new thread on islands of function.

If you would like to start a thread, let me know. I think I am able to give you that ability.

I’m not sure, but I think you can contact me with the mail link on the front page of TSZ. Alternatively, click on my name to get to my blog, hold out mouse over the “about” tab near the top of the page, and you should see a “contact me” link.

41. If gpuccio’s basic argument is about “islands” of function, then his dFCSI concept has nothing to do with SI or CSI or FC. These concepts developed out of Leslie Orgel’s SI. They involve a scale of some sort (the canonical example would be fitness), and the idea of CSI is to define some region far enough out on the scale that pure mutation, unaided by natural selection, cannot be expected to get you there even if every particle in the universe was an organism, mutating as fast as particles interact.

So the argument is then that something nonrandom must have gotten the organism there. Then there were supposed to be theorems showing that natural selection could not have done that, leaving Design.

What has the whole argument about “islands” got to do with that?

42. This is becoming really boring. I have given very clear definitions of what NS and IS are.

It’s a shame that you find consistency so boring, because science depends on it. If you want your argument to be taken seriously, it needs to be consistent, and that means correcting your inconsistencies when they are pointed out to you.

The correct concept is as follows: It is completely wrong to model NS using IS, because they have different form and power.    [emphasis yours]

Now you have reversed yourself:

So, what about a GA that models NS? It is a model which implements parameters appropriate for what is being modeled. It will use intelligent selection, but giving it a mathematical form which mimics true natural selection as we can observe it in nature. In that sense, it can give useful information.

Those statements contradict each other. Which do you affirm, and which do you retract?

43. Joe F.:

So the argument is then that something nonrandom must have gotten the organism there. Then there were supposed to be theorems showing that natural selection could not have done that, leaving Design.

What has the whole argument about “islands” got to do with that?

Many (perhaps most) of the IDers concede that natural selection is effective, but they claim that its effectiveness is limited to microevolutionary change.

In other words, they think:

1) that organisms are spread across multiple “islands of function”;

2) that movement within each island (i.e. microevolution) is possible via mutation and natural selection; and

3) that movement from island to island (i.e. macroevolution) is impossible via mutation and natural selection, because the islands are too far apart. Therefore the Designer operates an inter-island ferry service.

There are many problems with this argument, of course, but that’s what they’re claiming, and that’s why gpuccio agreed with me when I wrote:

Gpuccio is therefore betting the farm on the hope that actual biological fitness landscapes will consist of separated “islands of function” which NS cannot navigate.

44. 3) that movement from island to island (i.e. macroevolution) is impossible via mutation and natural selection, because the islands are too far apart. Therefore the Designer operates the inter-island ferry service.

Aack, this is seriously misleading mostly because it embodies the creationist picture of “macroevolution” as one current “form” morphing somehow into another current “form”.

They envision these islands as having been poofed into existence at creation. We DO note that no current form morphs into another current form, so we do not observe macroevolution. This model doesn’t lend itself to the notion of islands splitting into two or more islands and drifting gradually apart, until the distance between them can no longer be bridged.

Evolution is the process of making new islands.  Not travelling from one existing island to another.

45. OK, thinking about it a bit more, my most recent comment was ill-thought-out. To answer my own question, I think the relation of SI, FC and CSI to gpuccio’s “islands” is that we make our scale the extent to which the particular function of these sequences is achieved. The distance out on that scale would be big enough if that much of that sequence could not be achieved by mutation alone. So it sort-of is related to Dembski’s argument.

Still, there is no proof that natural selection cannot get that far. And we know how gpuccio responds to this — he declares that it is an empirical generalization to which no exception has yet been found, and that no proof is needed.

46. And we know how gpuccio responds to this — he declares that it is an empirical generalization to which no exception has yet been found, and that no proof is needed.

So I guess it is not possible for more energetic transitions or tunneling to improbable states to occur in physics and chemistry. All biological creatures are tightly bound solids that no amount of energy or tunneling can change.

Oh dear; chemistry never happens! How did the entire scientific establishment miss the polls that have been saying that all along?

47. A common feature of Creationist argumentation is an inability to distinguish the continuous and the discrete. Finding the underlying pattern involves trying to go behind one’s own perceptions of how the world presents itself to us. I wonder if they think that the rainbow really is divided into 7 bands, rather than the physical interaction of photons of different wavelengths with our restricted set of opsins, and the subsequent brain processing …

One finds this in the certainty that, variously, species cannot change at all, or they cannot change beyond a certain range within a common morphology, or that a protein has a ‘function’ and can wobble about in that discrete patch but never outside it.

One could change one letter of a book every 1000 years. At no point does the text become anything other than a minor variant on what went before. And yet cumulatively, over say 100 million years, 100,000 substitutions have taken place, enough to obliterate the original. Books are islanded, of course. They must be comprehensible to be ‘viable’. And so in comes that other Creationist favourite, the argument from analogy. The space of all books is islanded, so the space of [string-system X] must also be islanded. One can therefore (fallaciously) eliminate incremental progression (including that concentrated by differential fitness, the ‘functional/specified’ part) as a ‘necessity mechanism’.

48. Languages present the best analogy I can think of for evolution. They are semiotic, their elements are discrete, their implementation is always physical.

Yet they evolve. we have written histories showing their evolution in great detail.

And we have examples like Basque, that are isolated and have no known historical connection to other languages.

I understand that languages are used by “intelligent” agents, but I have not seen any intelligence directing their evolution. Actually the reverse. Intelligent agents appear to oppose their evolution and take active steps to hinder their evolution.

49. The use of language turns out to be an important tactic of the ID/creationists.

Those who can recall the heyday of the “scientific” creationism vs. evolution debates back in the 1970s and 80s will remember the most common settings and the atmospheres in which those debates took place.

They were often sponsored by local “Creation Science Associations” – these sprung up all over the country – and the audiences were bussed in from surrounding churches to support the creationist debater.  They hooted and cheered in all the right places when the creationist debater evoked his practiced, sneering jabs at evolution.  It was pure political theater.

The rules of engagement in these debates were usually written by the creationists, and they usually included a rule forbidding the “evolutionist” from bringing up the connection between “scientific” creationism and sectarian beliefs.

The other main feature of these debates was that the creationists insisted on the language to be used.  Every taunt and every critique of science was couched in the language of the creationists.  Their opponents were expected to answer using creationist concepts and language.  Any attempts on the part of the opponent to correct the concepts and language would be met with some kind of implication that the person, even though an expert in the science, didn’t understand science and the methods of science.  Gish and Brown used this tactic quite agressively.

ID/creationists are still using these tactics today on their websites.  That UD website is a classic example, as are the AiG and ICR sites.  Over at UD they appear to have recruited the hecklers also.  There are at least two flying monkeys over there who are constantly flinging feces at anyone who takes issue with any of the big gurus.

All the language over there is made up ID/creationist language.  They have their own “information”, their own “function,” their own “entropy,” their own “second law of thermodynamics,” their own “kinds,” their own rules about atoms and molecules, their own “mathematics,” their own “fitness landscapes;” no matter what it is, they have their own version of it and expect their opponents to speak in their language.  They don’t acknowledge or accept any of the concepts and language developed around science.

So anyone who wants to debate them is expected to answer in their language and put up with their hecklers.  Answers that use concepts from science are not recognized or acknowledged.  Whenever any ID/creationist opponent makes a good point, they are immediately bombarded with the Gish Gallop in the form of huge dumps of copy/paste junk along with a condescending scolding.

Not only is ID a child of “scientific” creationism by political design in order to get around the courts, it contains all the same language, the socio/political tactics, and the same sets of fundamental misconceptions and misrepresentations.

And now they are trying to use the Gish Gallop to distance themselves from “scientific” creationism while claiming that they intellectually legitimate with no sectarian motives.

50. Mike Elzinga has pointed out that UD commenters have their own definitions of many terms. I’d add “model of evolution” as another one that they have assumed that they can redefine.

51. If gpuccio really is able to identify dFSCI, then I would ask him to consider the following experiment:

First, examine the dog genome for obvious examples of dFSCI. I should think that ID predicts, implicitly, that there ought to be at least one example unique to dogs.

Second, examine the genome of a species widely believed to be of another “form,” yet close to the dog lineage (at least by evolutionary thinking), like that of the domesticated house cat. Again, there ought to be at least one unique example.

Third, use the dFSCI data from the previous steps to determine the status of an animal that may or may not be an intermediate form, like that of true foxes — is Vulpes a product of “micro-evolution” or of the Intelligent Designer(s)?

Not only would this be a practical example of dFSCI’s ability to detect design, but it would also be the first step towards an ID version of Linnaean taxonomy — a map of the actual islands of functionality found in nature.

52. I’d add “model of evolution” as another one that they have assumed that they can redefine.

Yes indeed!

Henry Morris and Duane Gish hammered on this from the beginning. Evolution was defined as organisms getting better and better and better; progressing upward toward “higher” states, toward more perfection, and toward lower entropy.

Then Morris threw in the ultimate refutation of “evolution” – as defined by the creationists – by redefining entropy and the second law of thermodynamics and asserting that this “fundamental law of the universe” directly refutes “evolution” (as defined by creationists).

They blasted this at biology teachers and at the general public. I have in my files multipage newspaper clippings of the generous coverage given to creationists by local newspapers at the time. In these newspaper articles creationists – claiming to have doctorates of some sort – would be laying out all these definitions and carefully painting the contradictions with their charicatures of science.

They “uncovered the hidden skeletons in the closets of the science community;” and with copious use of inuendo, they labeled the scientific community as a bunch of deceivers. The books and booklets published by the ICR were the classic, paranoid diatribes against the “dark secrets” of evolutionists. Evolutionists were “exposed” admitting that they didn’t really believe there was any evidence for evolution.

Everything coming out of the creationist movement was fabricated. My impressions at the beginning of all that were that it was so stupid that nobody would take it seriously. I was wrong; as were many others in the science community. We didn’t know how extensive or organized it already was. We were simply very naive about those kinds of socio/political tactics.

I still look back and shake my head at the brazeness of Duane Gish harrassing teachers in front of their students. Schools were much more open back then; and Gish would simply latch onto a fundamentalist student to “invite” him to visit their biology class.

53. Mung (#949) doesn’t “get it”:

[me:] Take a population of one million mosquitos. If allele A ot one locus has a gene frequency of 0.0001, and allele B at another locus has a frequency of 0.0001 also, then if they are associated at random, the haplotype AB would basically not exist in the population, as it would have an expected frequency of 0.00000001. Now suppose that A and B are favorable. Each rises to a frequency of 0.01. Now recombination between these loci would create AB haplotypes at a frequency of 0.0001, which is high enough that they really would exist in the population.

[Mung:] Assume that as ‘A’ increases in frequency ‘a’ decreases in frequency. Assume that as ‘B’ increases in frequency ‘b’ decreases in frequency.

So while you have increased the probability of AB you have decreased the probability of ab.

And the probability of Ab and Ba?

I still understand why you think this is some great “probability increaser.” It’s not.

However dumb I may be, even I realize that the four haplotype frequencies have to add up to 1. So you can’t increase all of them at the same time.

The issue was whether a new type could come into existence as the result of natural selection. These are haplotypes. In the original population there were 1,000,000 mosquitos and an expected frequency of the AB haplotype of 0.00000001, which means basically no AB’s at all. After gene frequencies of A and of B increase by natural selection, now there can very easily be AB haplotypes. So we have answered the question, and you are wrong: natural selection creates the conditions for AB to exist.

(To anticipate the usual objections:) Yes, recombination is involved too, but without the natural selection it would not make AB’s. The semantic quibble that natural selection doesn’t do it without another evolutionary forces such as recombination is that, an irrelevant quibble.

And yes, the frequencies of some of the other haplotypes will decline. That is very sad for them, but this will always happen when the population’s mix of genotypes changes.

Finally, yes, it is merely a matter of probabilities. Population genetics kind of tends to be that way.  :-)

So Mung’s (#931) assertions that because natural selection both increases and decreases frequencies of genotypes, it’s not special, and by implication can do nothing are wrong. When land vertebrate forelimbs evolved from fins, the frequencies of alleles that made fins decreased as the frequencies of alleles that made limbs increased, and that is not a problem for evolution.

54. Mung has a “gotcha” moment:

Me: A common feature of Creationist argumentation is an inability to distinguish the continuous and the discrete…

One could change one letter of a book every 1000 years.

Mung:That would be a discrete change. What’s your point?

Link two sentences somewhat separated in my original post. Point out that genetic change is digital at the lowest level and – bingo! – my point is neatly obfuscated by definology, or so Mung appears to believe, though he rather helps to illustrate it (and Mike’s, that word gaming is at the heart of much of the argumentation). I’m sure there is no such thing as a continuous sound generated from an MP3 file either, if you had responsive enough speakers and ears. And let’s not forget quanta!

My continous/discrete distinction related to biological categories, of which I provided examples – species, ‘kinds’, or protein function, sets around which we can attempt to draw our wiggly Venn-diagram lines. Incremental genetic change can (on the evolutionary paradigm) move populations hither and thither. The wiggly lines aren’t fixed – we, categorisers, draw them. If we lived long enough, we’d have to keep redrawing them. The Creationist, however, sees these lines as indicative of some fundamental essence: ‘macro-discrete’ categories between which Mung’s pedantically ‘micro-discrete’ genetic change cannot travel.

55. I wanted to discuss one more detail which will end up being one more reason why this thread has run its course. It had been suggested (by patrick on December 4 at 10:29pm above) that the `tierra` system would be a simulation of evolution by natural selection, mutation, and some other genetic forces that might meet gpuccio’s conditions. Thus it could be used to see whether dFCSI would arise in such a simulation.

But there is a problem. dFCSI requires an assessment of “function”. tierra has no clearly definable function other than survival — whether a particular genotype persists. I cannot easily see how we could use that to assess “function” for dFCSI. Survival is assessed purely empirically by whether the genotype comes to take over the tierra “world”, or whether it persists in it. That cannot be evaluated prospectively. One cannot compute fitness by examining the genotype and using a table of genotype fitnesses (as one can in more conventional population genetics models of evolution). Thus there is no way to know whether the fitness is very high just by looking at the genotype.

So this suggestion will probably not fly. gpuccio might (wearing the “Oracle of Naturalness” hat) declare it sufficiently “natural”, but functionality would probably prove impossible to assess in a way that gpuccio would approve.

gpuccio has seemingly also ruled out GA-type models (although keiths has pointed out contradictory statements gpuccio has made on this point, and there is as yet no clarification of the matter by gpuccio).

So we seem still to be in the situation that there is no tractable model that gpuccio would approve of that could be used to investigate whether gpuccio’s claims are plausible.

56. Mung (UD Jerad thread #987) has continued to try to (mis)characterize natural selection as ineffective. My comments were directed toward one issue: whether natural selection could bring about the appearance of new combinations of alleles, the example being a haplotype AB.

The example I gave did require recombination to be present, to put together the two alleles once they are present in high enough frequency to make this event probable (another possibility involves mutation at one locus once the allele at the other is frequent enough). Yes, of course it is possible for that event to occur when A and B are very rare — it is just very improbable for AB to exist in the population in the example I gave. Mung had made it sound as if natural selection had no effect on whether new combinations of alleles would occur. It would have a big effect in this example. So Mung was wrong about that.

As I have explained above, I made no assertion that this meant that all haplotypes increase as a result. I explain above that if some haplotypes become less frequent, that is no problem. I did not use the term “combinatorial probability generator” nor do I think that is a useful phrase.

The rest of Mung’s comment is word games. The issues are not those Mung imagines, and I have no interest in spending time on those.

57. Mung has corrected me (in UD Jerad thread #988) on my wording on two points: I should have said “survival and reproduction” instead of just “survival”. And yes, the “function” needed is the function of a digital organism within tierra, not the function of the whole simulation.

If there is some other “function” Mung has discerned for the genotypes in tierra, it would be interesting to know what it is — I don’t see it.

Perhaps when gpuccio has looked more closely at tierra, we can be told whether there is any other “function” available in tierra, and gpuccio can clarify whether or not a simulation like tierra can be used to assess dFCSI. Until that is clarified there is little more to say.

58. One of the more interesting features that arise in some tierra runs is parasitism followed by hyperparasitism.  I wonder if those qualify as functions.

As you say, though, any discussion is blocked pending clarification from gpuccio.

59. Actually Joe, I was thinking of Avida and not Tierra, but I got lucky, lol. After all, what’s an evolutionary simulation without reproduction!

I need to take a look at Tierra specifically. I haven’t seen gpuccio in a few days =p. I hope he’ll return.

Is there no scenario in Tierra according to which some organisms leave more offspring than others?

Do you consider Tierra to be an evolutionary simulation? If not, why not? Is it because you can see no implementation of natural selection?

60. Well, I believe I have answered Keiths’s comment in my post #941.

You still haven’t resolved the contradiction. I responded to your #941 here:

keiths on December 6, 2012 at 1:19 am said:

This is becoming really boring. I have given very clear definitions of what NS and IS are.

It’s a shame that you find consistency so boring, because science depends on it. If you want your argument to be taken seriously, it needs to be consistent, and that means correcting your inconsistencies when they are pointed out to you.

The correct concept is as follows: It is completely wrong to model NS using IS, because they have different form and power. [emphasis yours]

Now you have reversed yourself:

So, what about a GA that models NS? It is a model which implements parameters appropriate for what is being modeled. It will use intelligent selection, but giving it a mathematical form which mimics true natural selection as we can observe it in nature. In that sense, it can give useful information.

Those statements contradict each other. Which do you affirm, and which do you retract?

61. Gpuccio,

You also haven’t responded to this comment:

keiths on December 5, 2012 at 1:48 am said:

To be able to explain that by a designer (the only credible explanation available) I have to assume that the designer worked in a way that explains the nested hierarchies.

And in particular, the objective nested hierarchy. The problem is that you know nothing about the designer, so you can’t independently justify your assumptions. He’s an unknown designer (maybe more than one), with unknown abilities, unknown limitations, and unknown goals.

The only “justification” you offer for your assumptions is that you want to fit your hypothesis to the evidence of the objective nested hierarchy. But then you’re committing the Rain Fairy fallacy.

It’s the equivalent of this:

You are arguing against a Rain Fairy proponent. You point out that under the Rain Fairy hypothesis, there is no reason to expect low-pressure systems to rotate in opposite directions in the two hemispheres. Any wind pattern is compatible with the Rain Fairy hypothesis, so it is completely arbitrary to assume that the Rain Fairy just happens to choose (or is somehow limited to choosing) the counter-rotating scheme.

Your opponent responds, “I am making a simple and reasonable assumption. I assume that the Rain Fairy likes symmetry, and this is why low-pressure systems rotate in opposite directions in the two hemispheres.”

Are you persuaded by the Rain Fairy advocate? If not, then why should we be persuaded by your argument, which is logically identical?

To summarize, the problem is that if you know nothing about the designer, you can’t independently justify any assumptions you make about him. And if you try to force-fit your hypothesis to the evidence by simply tacking on arbitrary assumptions, then you are committing the Rain Fairy fallacy.

62. My statement was:

“The correct concept is as follows: It is completely wrong to model NS using IS, because they have different form and power.”

That is absolutely true for all the GAs you guys propose, which are based on IS and try in no way to realistically model NS.

In other words, you would like to change the meaning of your statement without admitting that you are changing anything.

The correct concept is as follows: It is completely wrong to model NS using IS, because they have different form and power.   [emphasis yours]

You want your statement to mean this:

It is wrong to model NS using IS unless the model is realistic.

Your refusal to admit your mistake is just silly, but can I at least get you to agree, for the sake of future argument, that the revised statement is correct?

63. I am really curious about the power of itnelligent selection. I have asked for several years why gpuccio thinks IS is superior. I’m specifically interested in how IS would apply in the Lenski experiment.

What I want to know is how the intelligent selector  knows to favor neutral precursor mutations, those that confer no immediate reproductive advantage but which enable later mutations to become adaptive.

How does this work?

64. Keiths:

I will not answer your “argument” about the Rain Fairy. I find it simply stupid, with all respect.

Gpuccio,

You won’t answer it because you can’t answer it. Pretending it’s stupid is just a way of saving face.

You’re stuck between a rock and a hard place:

1a. Unguided evolution is far better than ID at explaining the evidence of the objective nested hierarchy.

1b. Meteorology is far better than the Rain Fairy hypothesis at explaining the weather.

2a. The Designer is an unknown being with unknown abilities, unknown limitations, and unknown goals. ID therefore predicts nothing, and can be fitted to any set of facts about life by simply saying “that’s how the Designer did it.”

2b. The Rain Fairy is an unknown being with unknown abilities, unknown limitations, and unknown goals. The Rain Fairy hypothesis therefore predicts nothing, and can be fitted to any set of facts about the weather by saying “that’s how the Rain Fairy does it.”

3a. To bring ID into alignment with the biological evidence, you have to make a bunch of assumptions about how the Designer operates.

3b. To bring the Rain Fairy hypothesis into alignment with the meteorological evidence, you have to make a bunch of assumptions about how the Rain Fairy operates.

4a. There’s no independent justification for the assumptions you add to the ID hypothesis. You’re just forcing ID to fit the evidence. All the work is being done by your arbitrary assumptions, not by the theory itself.

4b. There’s no independent justification for the assumptions you add to the Rain Fairy hypothesis. You’re just forcing the Rain Fairy hypothesis to fit the evidence. All the work is being done by your arbitrary assumptions, not by the theory itself.

Where does all of this leave you? You’re in the embarrassing position of either

a) supporting both ID and the ridiculous Rain Fairy hypothesis, or

b) admitting that unguided evolution fits the evidence far better than ID, just as meteorology fits the evidence far better than the Rain Fairy hypothesis.

Rather than facing up to this, you’ve chosen to avoid the dilemma by pretending that the argument is stupid — and hoping that someone will believe you.

65. keiths: Gpuccio,

You won’t answer it because you can’t answer it. Pretending it’s stupid is just a way of saving face.

You’ve stated the case very well.

ID doesn’t bring anything to the table in the way of mechanisms and is thus useless as an aid in understanding why life responds to the environment it finds itself in.

ID also doesn’t answer another important question and that is, “How does the designer know what will be required in the future?”

66. kairosfocus weighs in on the Rain Fairy argument. He predictably makes multiple references to strawmen, then just as predictably proceeds to set up and topple his own strawman.

(And, on the empirically observable sign FSCO/I in its various forms including dFSCI, it is abundantly confirmed to be reliable at that empirically with billions of cases in point. That is, there are no credible false positives for FSCO/I beyond 500 – 1,000 bits…

In other words, if you assume that evolution isn’t a credible explanation of dFSCI, you will conclude that evolution isn’t a credible explanation of dFSCI.  Take that, evolutionists!

So, while KS et al do not wish to accept it, we have a highly reliable sign that is best explained on the known and observed causal factor, design.

We know that design can produce functional complexity, but that doesn’t justify the conclusion you’re leaping to: that only design can produce functional complexity. Where’s your evidence?

I’ve shown that unguided evolution explains the evidence literally trillions of times better than intelligent design. I even challenged you to respond, placing absolutely no restrictions on the venue, format, or length of your response. You’ve been evading the challenge ever since. Why is that?

Now, the next thing is that a strawman demand is set up that a designer of life, to be acceptable to KS et al, must start from scratch every time, instead of using and even modifying a code base.

I’ve made no such demand. You would know this if you had bothered to familiarize yourself with my argument before presuming to criticize it.

I take issue with Gpuccio’s assumption about design reuse not because I think he should have assumed the opposite, but because I think he is not entitled to assume anything at all, unless he provides independent justification for his assumptions. Otherwise he is committing the Rain Fairy fallacy. See this comment.

ID can be made to fit design with reuse, or design from scratch. It can be made to fit modular design, or monolithic design. It can be made to fit anything at all, by simply stipulating that the Designer must have done it that way. ID is infinitely malleable. It fits anything, which means that it predicts and explains nothing. That is its great weakness.

Evolution is the better theory.

67. petrushka,

I am really curious about the power of intelligent selection. I have asked for several years why gpuccio thinks IS is superior.

Think WEASEL. Think Genetic Algorithms.

68. Toronto:

ID doesn’t bring anything to the table in the way of mechanisms…

And yet you managed to post on this site. No intelligent design required.

ID also doesn’t answer another important question and that is, “How does the designer know what will be required in the future?”

The people who designed this web site had no idea what you would post here on December 13, 2012. How on earth did they manage to design for what would be required in the future?

69. Mung: “The people who designed this web site had no idea what you would post here on December 13, 2012. How on earth did they manage to design for what would be required in the future?”

You seem to be on our side again with your analogy.

The website designers foresaw the need to host comments and designed a site that would accept a combination of ASCII characters to represent those comments, but they did not require any knowledge of what the actual configuration of those ASCII characters in my comment would be.

Tell me what the weather will be like in a hundred years

There is no spec anyone could write that would give us enough foresight to modify the “information” in every single biological design that would require change to exist in that unknown future environment.

How does the designer know what to change for an unknown to him, future environment?

How many organisms would be affected by this change?

Who will be affected more, herbivores or carnivores?

Should he hedge his bet and make 70% of the organisms herbivores to balance out predator/prey relationships?

70. There is no spec anyone could write that would give us enough foresight to modify the “information” in every single biological design that would require change to exist in that unknown future environment.

So?

How does the designer know what to change for an unknown to him, future environment?

There’s no rule of design that demands that designers must plan for all possible future contingencies. Our universal experience with known designers indicates that there is no such requirement.

71. Explain what WEASEL is in your own words, and what it is intended to demonstrate.

WEASEL is a computer program written by Richard Dawkins as an exercise in demonstrating the “power of cumulative selection” in which strings of characters are copied and mutated and then compared to a target phrase with those strings which more closely resemble the target phrase being selected to seed the next round, repeating the process until a string exactly matching the target phrase is found.

There are various other versions of it in many different programming languages freely available on the internet.

It’s a fine example of the power of intelligent selection, though that’s hardly what Dawkins intended to demonstrate with it.

When your side comes up with a version of it that uses natural selection, rather than intelligent selection, we can compare the differences in “power” and answer petrushka’s question.

72. Give us an operational definition of natural selection, so we will know what to shoot for.

73. Mung: “So?”

Design me a program.

Here’s what it has to do: ………………..

Your code is a critical component so let me know right away if you can’t meet any part of the spec.

74. Toronto:

Here’s what it has to do: ………………..

Program 1:

puts “………………..”

Program 2:

20.times {print ‘.’}; puts

Program 3:

def print_periods how_many

how_many.times {print ‘.’}; puts

end

Program 4:

def print_char chr, how_many

how_many.times {print chr}; puts

end

Your code is a critical component so let me know right away if you can’t meet any part of the spec.

First we’ll need to work together to write tests, that way we’ll know when the program has met your specification(s).

75. Mung: First we’ll need to work together to write tests, that way we’ll know when the program has met your specification(s).

76. The intelligent designer had better know what’s coming in the future and prepare for it.

If he doesn’t, he’ll be modifying our “code” 24/7.

In that case, science would be impossible since the intelligent designer will be interfering with natural processes on a daily basis.

Designers of fighter aircraft can’t ignore a future that’s decades away and neither can the intelligent designer.

77. Designers of fighter aircraft can’t ignore a future that’s decades away and neither can the intelligent designer.

What intelligent designer?

How about a designer of disposable razors or cotton swabs?

In that case, science would be impossible since the intelligent designer will be interfering with natural processes on a daily basis.

You don’t know that they aren’t. How would you know?

78. Toronto: In that case, science would be impossible since the intelligent designer will be interfering with natural processes on a daily basis.

Mung: You don’t know that they aren’t. How would you know?

That’s why ID is a non-scientific concept.

79. You don’t know that they aren’t. How would you know?

That. in a nutshell, is why ID is useless.

Once you have made the assumption that natural processes are capricious, you have abandoned any hope of doing science, because science is the business of finding regularities. The short answer is that you cannot be certain that nature is not capricious or that the fabric of reality is manipulated by the matrix masters.

But based on the last couple hundred years, that’s the way to bet. When theists get sick, most will go to a doctor who relies on reductionist scientific medicine.

EDIT:

All the sciency machinery of ID is engaged in the hunt for discontinuities. Gaps, if you will. Miracles by another name.

All the work and writings of Axe, Dembski, Behe, et al, are deployed to find events and phenomena that cannot current be explained. I Some argue that there is some utility in that, because it sometimes suggests areas that need to be studied.

But it is highly inefficient, because science goes to those places anyway. And it is entirely ineffectual, because unexplored areas shrink.

80. Stumbled across the following post by gpuccio:

Intelligent selecion is a powerful principle, as shown in bottom up protein engineering.

There are two fundamental differences between IS and NS:

a) IS can select for any defined function, even if not immediately useful. NS can select only for those functions that give a reproductive advantage in a specific context (that is, an extremely tiny subset of all possible functions).

b) IS can select functions even at very low levels. IOWs, IS can recognize a function even in its raw manifestation, and then optimize it. NS requires that the function level be high enough that it can give the reproductive advantage at phenotipic level.

Both points are extremely important, and both points are the consequence of the intervention of intelligence and purpose in the process. Moreover, bottom up IS can well be integrated with top down engineering in the design process.

All those possibilities are denied to non intelligent processes.

81. petrushka:

Gpuccio is arguing that because no one has completed the painstaking research to find the history of protein domains, such histories cannot exist.

That is incorrect.

The painstaking research is in. That’s what allows us to identify the protein superfamilies. It is that same painstaking research which allows us to say that the intermediates are missing.

82. It is that same painstaking research which allows us to say that the intermediates are missing.

Someone has done research and concluded that billion-years-dead genomes are no longer available to sequence? I wonder if I could get a slice of that funding?

83. Allan:

Someone has done research and concluded that billion-years-dead genomes are no longer available to sequence? I wonder if I could get a slice of that funding?