Tuesday, January 15, 2013

The Mathematics of Intelligent Design

Here we have a classic example of why it's hard to take intelligent design seriously.

The flagship blog of intelligent design presents a worthless piece of software, not even at the level of a bad junior-high-school science project, as an accomplishment. It `tries to answer questions like this: “a random process generating sequences of length L from an alphabet of S symbols in T trials of t seconds each, involving c chemical reactions, does exceed the resources of the universe (age, max number of chemical reactions, universal probability bound)?”'. We are told that this silly exercise "may give an idea of the numbers involved in scenarios as origin of life, production of biopolymers, binary and character text generation, and so on." Right.

The author clearly doesn't know what "random process" means (hint: it doesn't necessarily mean uniform probability). And his program doesn't take into account anything interesting about chemistry at all. It's just worthless number pushing.

Garbage in, garbage out. Come to think of it, that's pretty much the description of intelligent design.

Addendum: they've already removed the page. I guess there are some things that are so stupid, even Uncommon Descent can't get behind it. But you can still see the software here.

Addendum: it's now back again. Not much different than before, except they added a few English mistakes.

23 comments:

nwrickert said...

That UD post disappeared about as fast as it appeared.

It showed up in my RSS reader, but the post had already gone missing by the time I tried to open it.

No I am wondering whether it went missing because it was so stupid, or whether it went missing because some of the default parameters offended YECs.

Diogenes said...

Whatever happened to Dembski's "phi" factor for descriptive simplicity?

SPARC said...

According to the "-2"-ending of the URL in your link to the article you were already dealing with the second version of the post. The current URL ends with "-6". Thus, niwrad must have been working really hard on it.

Diogenes said...

Look, let's attack. Let's use their handy calculator to prove everything is intelligently designed.

1. Sodium chloride crystal
2. Iron oxide (rust)
3. Oil and water after separation
4. The Christian God

Also compute CSI before and after natural processes:

1. Short gene (microRNA) before & after gene duplication
2. Oil and water before &after separation

Prove CSI increases naturally. Also, Christian God intelligently designed.

Muskat said...

Saying that "random process" definitely means "uniform probability" seems so silly, that I almost need proof that Newt... I mean Dembski, (or niwrad) actually said it.

Anonymous said...

I guess there are some things that are so stupid, even Uncommon Descent can't get behind it.

I guess this hypothesis has been falsified.

TomS

RBH said...

Diogenes suggested that we run the Christian God through the probability calculator on UD. Years ago I suggested that if one runs the Christian God through Dembski's explanatory filter one reaches the inescapable conclusion that God was designed. I agree with that conclusion: God was designed by humans. Wesley Elsberry reached the same conclusion here in 2003:

I think all believers would agree that God did not arise from either regularity or chance. However by Dembski's definition of design only regularity, chance and design are allowed -- Dembski defines design as everything which is neither regularity nor chance. With this definition of design it is inevitable that the Explanatory Filter decides that God is designed.

Given the wider aims of the ID movement it is amusing to see that Dembski's Explanatory Filter appears to give support to the atheist argument that God was designed by humans.

Bensenhurst said...

"Dembski defines design as everything which is neither regularity nor chance."

Is that precise? Wouldn't he define design as everything material which is neither regularity nor chance?
If so, that nullifies your entire clever argument.

Muskat said...

I'm calling Shallit's bluff. Where does niwrad say, or imply, that "random process" definitely means "uniform probability"?

Tim said...

Disclaimer: this is a mathematical tool dealing with processes considered purely random. It doesn’t pretend to simulate the physical, chemical, biological, functional constraints specific of the processes.

What a bizarre confabulation. How does this engage evolutionary theory in any way?

Jeffrey Shallit said...

Where does niwrad say, or imply, that "random process" definitely means "uniform probability"?

He doesn't say it explicitly, but it is implicit in his model. But then, I am used to dealing with people with poor reading comprehension skills. Did you take a look at his web form? There is nothing it in that allows non-uniform probabilities of events.

Diogenes said...

He doesn't say it explicitly, but it is implicit in his model.

Jeff is right. Niwrad gives the equations for his probability of a sequence: P1 = S^-L where S = size of alphabet, L = sequence length.

This equation is only true if the probability of all "letters" in the sequence are equal and independent.

Then p for one letter p = 1 / S. That's the flat probability part. Assume all letters in the sequence are independent, then P = p^L = S^-L as Niwrad thinks.

As I like to call it, the tornado probability. Has nothing to do with evolution or any observed process, including tornadoes.

Diogenes said...

Jeff: I am used to dealing with people with poor reading comprehension skills.

You said it, brother.

Muskat said...

I appreciate, and accept your answers. The reason I was so pushy was because I asked this same question a long time ago at this site (when Jeff wrote the same thing), but never got an answer. Now I got my answer.

Jeffrey Shallit said...

Muskat, we (Elsberry and I) addressed the question of uniform probability in Dembski's work in detail in our long paper published in Synthese. You can find the even longer version of it easily online at talkreason.org.

Rumraket said...

One of the issues with silly ID calculations like these is that IDcreationists presuppose that evolution had to find some single and *specific* sequence to get a functional biological polymer, instead of just anything that works.

They can never seem to wrap their heads around the idea of contingency. That evolution didn't really do a seach with a specific goal in mind, what happened was just that genes mutated over generations and sooner or later, something stuck because it happened to work in the requisite organism.

That given a different environment and an organism with a different evolutionary history, something else would have evolved that would have fit that different organism, in it's different environment, just as well as this extant protein does now.

But to IDcreationists, they're sitting at the end of this long sequence of events and and thinking because something that works resulted from it, it must have been designed.

It's a sort of texas sharpshooter fallacy, mixed in with the unsubstantiated(and indeed counterfactual) belief that functional biomolecules are extremely rare among the total set of possible combinations.

As if there was no evolutionary pathway to some extant protein, because they believe evolution had to search through some enormous sequence-space of nonfunctionality to get that one and specific extant sequence, and that without that one specific sequence, no other molecule would have been functional in it's stead.

Contingency, contingency, contingency. They'll never get it. They resulted from it, and they just cannot fathom that this could have happened without design, planning and foresight.

Diogenes said...

@Jeff,

I myself am having trouble finding your anti-Dembski paper at talkreason. Their search engine appears not too bright.

Diogenes said...

@Muskat,

I did not mean to imply you have poor reading skills.

I just mean that we are used to debating dishonest and ignorant opponents ;)

I think I found the Elsberry & Shallit paper: this 2004 paper.

Here are two good anti-Dembski papers by Elsberry:

IDiot Dembski makes bonehead programming errors, claimes to disprove evolution. (A must read.)

Elsberry demolishes Dembksi's moving goal posts

Jeffrey Shallit said...

Yes, Dio, that's the one.

Anonymous said...

Well, evolutionists also have their silly software programs, e.g. making dots in the shape of a spider and claiming that says something about natural selection!
Or maybe they could play the game people like you do, saying "I'm so much more intelligent than them, they couldn't possibly understand what I mean." Both of these are common in Darwinist arguments.

Jeffrey Shallit said...

Well, evolutionists also have their silly software programs, e.g. making dots in the shape of a spider and claiming that says something about natural selection!

I assume you are referring to Dawkins' biomorphs. I'd wager, however, that you have never actually read The Blind Watchmaker where these appeared. There they were used as an analogy of natural selection, with the person playing the role of the environment. They were not advanced as a tool to determine the feasibility of any particular evolutionary scenario.

Analogies are used all the time for pedagogical purposes. Today in class I used an analogy between finite automata and electric circuits, but it doesn't mean that automata are circuits.

As for your latter claim, I have never said, "I'm so much more intelligent than them", and you are a liar for claiming so.

Anonymous said...

Contingency, contingency, contingency. They'll never get it. They resulted from it, and they just cannot fathom that this could have happened without design, planning and foresight.

Yet everyone admits that contingency is a factor in our individual origins. Our genetics is the result of random mixing of our parents' genes, plus some random mutations. The history in which our ancestors happened to meet is full of events which were not planned or designed to generate us.

How about using this algorithm to compute the improbability of my birth on the assumption of reproductive biology, and thus prove Scientific Storkism?

TomS

Scepptic said...

"They'll never get it. They resulted from it, and they just cannot fathom that this could have happened without design, planning and foresight."
Meh, they'll just say that you cannot fathom that this could have happened with design, planning and foresight. Even if your argument is right, it'll never get you anywhere.