The design inference

[QUOTE]Originally posted by John Paul:
[b]It's called the design inference, not the materialistic naturalism excludance. [/QUOTE]Ok. Let's get one thing clear. Scientists and many others infer design in many ways. What Dembski is claiming is that
: he has identified a technique for inferring design which does not give false positives;
: this technique underlies design inferences that are made in other fields.Let's kick off with an interesting question regarding Dembski's claim to have formulated a logical filter for inferring design that does not give false positives. Is this filter now knowingly applied in any of the fields Dembski refers to as using design inferences? Let's take SETI and Forensic Science: now that Dembski has codified this tool is it actually being used in thse fields?This is not a rhetorical question. I don't know the answer and would genuinely be interested to see practical applications of Dembski's filter in real-world situations where decisions have to be made on the outcome. [b] [QUOTE]It is evidenced by the CSI and apparent IC of living organisms.[/B][/QUOTE]I can't imagine what you mean here. At the most this sentence can mean:If FW then (if LC then LD)
LC
Therefore FW.(Where F = "The Filter Works", LC = "Life is complex and specified", LD = "life is designed".)This is clearly bunk.BTW, I presume that having read Dembski, apparently approvingly, you don't mind me using completely superfluous quasi-symbolical forms in his manner? [b] [QUOTE]Mr. P has a point in that Dembski's filter can't detect random looking markings, but it sure could point out any accompanying text. [/b][/QUOTE]Well I wound't want it to detect them - I would want it to help me decide whether they were designed or not. [b] [QUOTE](Sorry Mr. P but DNA and the cell are hardly random markings that may or may not mean something)[/b][/QUOTE]O don't be sorry, just tell me how DNA is specified in Dembski's sense. [b] [QUOTE]All the filter wants you to do is consider the evidence and if you get to the 3rd box, use as much scrutiny of the evidence as technology allows before reaching a conclusion.[/B][/QUOTE]If this is the case then the filter is totally useless, because in the third box you are back to Paley's old argument that this thing looks like it might be designed, and Hume's rejoinder questioning what could be "sufficiently like" to justify the inference.Unless the filter can actually get you to the point of saying "this is designed" then it is as much use as the proverbial chocolate teapot.[b] [QUOTE]ID was shut out when the a priori before the black box was even opened.[/b][/QUOTE]I don't understand. Can you clarify? Thanks.[b] [QUOTE]Today we describe biochemical systems analogous with machines.[/b][/QUOTE]Analogy isn't going to do the trick, mate. What about the other way round? Let's say I compare the claw on the production line robot to a human hand. I remark how similar they are - how they have "fingers", "joints", how they move, how they grasp, how they have "tendons." I ponder on how they came about ... Does the hand help me accurately guess that the robot claw was made from cast and machined steel parts, that it was wired and soldered? Does the robot claw help me accurately guess that the hand was formed by cell division? Analogy clearly doesn't help in pondering their manufacture - why do you think analogy is going to get you any further when considering if they are "designed"?[b] [QUOTE] The way information is transported internal to each cell is analogous to a LAN (local are network- packeted, with header containing destination, source, key, data) with ports to the whole system, itself a myriad of complex pathways, in complex metazoans and other multicellular organisms.[/b][/QUOTE]Same thing. The human mind likes analogies because it enables us to reuse existing knowledge - but argument from analogy is illogical.
[b] [QUOTE]So if you are telling me that for some unknown reason, science has to exclude the design inference[/b][/QUOTE]Science does not have to exclude the design inference, it is purely that so far no one has come close to showing how it applies to living things in a way that helps scientists to understand them better than other methods. The really sad thing about Demsbki's work is that, rather like Goss's in the 19th century, it is irrelevant to the work being done in the field.(Which brings back to wondering if there are examples of his work being explicitly used in fields where it is of immediate relevance.)

​

[This message has been edited by Mister Pamboli, 02-28-2002]

[b] [QUOTE]What flaw? That never has anyone observed CSI arising via purely natural processes? Sorry, that's part of life. Bring us an indisputable example of CSI arising via purely natural processes and you will have exposed a flaw.[/b][/QUOTE]Here's an interesting one - lets see what you make of it.I refer to the miracle tomato of Huddersfield and the Bolton Aubergine. http://www.einterface.net/gamini/miracleislam.htmlIn these cases fruit was cut open and the patterns in the seeds and fibres spelled out quotations concerning Allah and His Prophet. The information was complex (unlikely to occur by chance), and specified (tractable information was there which could be easily read).Let's look at the probabilities involved to make sure the information is complex: first there is the probability of the fibres and seeds spelling out an Arabic phrase, presumably vanishingly small. Second, there are the probabilities that these phrases would refer to the Creator and His Prophet, presumably even smaller. Thirdly, the probabilities that these fruits would be bought my Arabic readers who could read the message. The probabilities are now surely enough to pass through the filter.Of course, there is the problem of whether the information is specified. It was certainly tractable - hundreds read the messages easily. The messages were certainly more tractable than the information in DNA. And, of course, the message had meaning and was significant.There are three potential outcomes to putting this through the filter:1: It is rejected by the filter because it is not complex. But it's difficult to see how these messages could occur with high probability. Even if the patterns naturally occuring in tomatoes and aubergines resemble Arabic writing we have to consider the probability of that occuring.2: It is rejected by the filter because it is not specified. For example you might say that the faitful only "imagined" they saw a message. But that doesn't wash, does it, because the entire point of the filter is to be able to distinguish "apparent" design from "real" design - not to give false positives. But the information in the tomato is clearly specified in Dembski's terms.3: The tomato and aubergine pass the filter. We have good reason to assume the messages proclaiming Allah and His Prophet are designed to be in the tomato.

​

[This message has been edited by Mister Pamboli, 03-13-2002]

quote:

---

Originally posted by Cobra_snake:
I will be the first to admit that I don't know what IC means. Could you please explain?

---

IC = Irreducible COmplexity. This is the idea that something may have a number of parts which make it work, but if any one of these is not present it cannot function. Behe's argument is (very very roughly) that such structures are typical of designed solutions and some of them could not have evolved but must have been created as "working modules" from the beginning.[b] [QUOTE]isn't it possible that aliens are completely different than life on earth, and therefore could of arised by natural processes?[/b][/QUOTE]So if it is possible to imagine the designer's life arising from natural causes, why can't we imagine our life similarly arising?The point made in the post is simple enough - if life is designed, a designer is implied. (Actually Dembski tries to avoid this conclusion, but not at all convincingly.) But all we are left with is who designed the designer?Your position is akin to asking "Who or what wrote Shakespeare's plays?" and accepting the answer "A pen."That's hardly the answer to life, the universe and everything, is it?

quote:

---

Originally posted by Cobra_snake:
"We may well see here the source of Behe's whole misguided campaign against Darwinism."

​

Interesting choice of words. "Crusade against Darwinism. It's not as though disagreeing with Darwinism is blasphemy. "

---

Hey Cobra - you don't normally trip up quite as quickly as that! Campaign in one sentence becomes crusade in another, though he doens't use the word crusade, and then you criticise the word that you introduced! Just how carefully did you read the article?

quote:

---

I suppose those people up at SETI should give up then.....

---

Why? How does this follow?

quote:

---

"Behe plainly admits that some cellular processes could have evolved by natural selection. If all those other cases didn't cause Behe to surrender his pet theory, why should one more?"
...
This statement is completely false.

---

In what way is this statement false? There is only one statement in the quoted passage - that Behe admits some cellular processes could have evolved. This is completely true - Behe does say this. (I know this because it was politely pointed out to me when I goofed over his current position on haemoglobin in another topic.)

quote:

---

Behe admits that some cellular processes could have evolved because they don't display irreducible complexity, which is the main argument of his book.

---

So, Orr's point is why does Behe think some are irreducibly complex and merely complex? The answer seems to be that where we have shown evolutionary pathways, Behe accepts them and it follows that the process cannot be irreducibly complex, but when we cannot show pathways Behe declares them to be irreducably complex. What he doesn't show is any a priori reasoning as to why some should be IC and others not, or any reasoning as to why IC structures should exist at all without an evolutionary pathway, or any means of deciding whether something is IC excpet only that we don't know the pathway.

JP, thanks for the answer. It's good to get involved in some debate again.You are right that Demsbki and some others are still developing his work in ID. No Free Lunch is basically an attempt to use some computer science work on genetic algorithms to suit his ID arguments - it's better written than his previous work, but still has the old Dembski failings: pretentious use of pesudo-formal logic and maths, a rather over-infalted sense of the importance of his own work (especially as he is borrowing ideas from others here) and the same fundamental flaw of never really comparing like with like. My advice? Read it, but don't buy it!My point about the use of his filter in SETI or forensic work is this: if, as he claims, his filter formalizes the underlying processes these fields already use, and if it does not give false positives, then do people use it? For instance, I design software development tools. Amongst other things, I look at the processes and patterns that programmers use in the real world and then turn those processes into software which helps them apply the techniques they already use, but more efficiently. If Dembksi's filter accurately describes the detection of design with no false positives, I would expect to see it being used to make the existing processes for detecting design more efficient.

quote:

---

When you say or imply that biological ID isn’t a scientific endeavor ...

---

I don't say that, but there are some that do. My problem with ID in biological science is the poor logic and the inadequate support for the inferences required to apply it.

quote:

---

Our ability to be confident of the design of the cilium or intracellular transport rests on the same principles to be confident of the design of anything: the ordering of separate components to achieve an identifiable function that depends sharply on the components. (Behe)

---

I like this quote because it sums up both the intuitive comfort that ID provides and also its fatal flaws.The intuitive comfort comes from the idea that we are simply using our common sense - we are detecting design in the same way as we detect the design of anything. The fatal flaw is this word anything because actually in other cases where we seek design we have contextual reasons to believe a designer is highly possible because we have seen very similar things being designed. Dembski and Behe's problem is to show that we can infer design in any context - and this they fail to do, because all their examples intuitively inferring "the design of anything" are highly contextualised. But organisms are a unique context, so unlike anything else in our experience that we cannot apply these techniques here with any confidence. Our inferences in other contexts are sound only insofar as the context lends weight to the inference.

quote:

---

What it can’t do is to tell you the meaning of the design. For example I could use the EF to determine that a Chinese character was designed but the filter wouldn’t translate that character for me.

---

Absolutely. I wonder if that will be Demsbki's next book - the meaning of life? My argument was that it couldn't identify what may or may not be writing as being designed and thus it couldn't be helpful.

quote:

---

From what we know about DNA not just any ordering of nucleotides will give rise to a living organism. The sequences that do allow for life would then be considered specified. Pretty basic actually.

---

There is a bit more to specification than that, is there not? I think Brachinus deals with this in another post. Remember that the filter also has to deal with all the junk DNA, retroviruses etc.

quote:

---

The EF is a basic process flow chart for inferring design. His book is called The Design Inference.

---

True. But Dr D does make a big point about the reliability of the inference. He goes to great lengths to explain his "magic number" which, if this is the probability of something occuring, and it passes through the filter, we can say absolutely that something is designed. Dembski is not saying "Hey this looks a bit like a designed thing, thinking of it in that way may lead to some interesting conclusions and lines of thought." He is saying "This is designed and that fact has important implications."If he was saying the former - that things may "look" designed and this can be a useful way of thinking of them to possibly find more information, then I would wholeheartedly agree. It is the inference that something "is" designed that I take issue with.Several of our later points dealt with the analogical nature of the design inference and its logical fallacy.Let's look at a really simple argument from analogy to see the problem, using Fingal, my old dog, as a subject: Fingal is like a cat, therefore Fingal is a cat.This is clearly wrong, but its premise is correct. Fingal is like a cat in many many ways. Thus the argument is formally a fallacy.So in the argument, "x is like a cat therefore x is a cat" the difficulty is clearly in the word "like": it has to have a very special meaning in the argument, referring to some very special attributes, to enable us to infer catness.Demsbki's design inference is basically "x is like a designed thing, therefore x is a designed thing." And for him, designed things do have some special attributes - complex sepcified information. His explanatory filter is a way of deciding whether something passes the like test. The argument remains formally fallacious, but could still be emprically useful if like can be determined accurately enough - this is what I meant by "Science does not have to exclude the design inference."THe difficulty remains whether these special attributes that Dembski claims designed things have, are sufficient for us to make this inference. No amount of logic can get round this problem, because the argument remains formally fallacious: the most Dembski can do is narrow the meaning of "like" to a point where we can make the inference with some confidence.So when you ask "Looks like a duck, waddles like a duck and quacks like a duck- what do you think it is?" we have to decide whether those criteria are sufficient for us to infer duckness. After all it could be an Egyptian Goose (Alopochen aegyptiacus). Or by your criteria my son's old electric toy duck (which sure as hell confused our cat!), or a cartoon duck. Your criteria aren't "sufficient" to infer real duckness. My argument is that Dembki's criteria aren't "sufficient" to infer design.

quote:

---

And just how has theorizing that life isn’t the product of design or Special Creation added anything of value to science?

---

It is the job of scientists to theorize. Every advance in natural sciences, including medicine is the result of investigating natural causes. Why should this investigation stop at the origins? Looking at the ultimate origins of life and how life has developed is an extension that examination of the processes of nature which has lead to every step of progress in the natural sciences. Can you think of any single tiny step of scientific progress which has been made by assuming the existence of a creator?

quote:

---

What is the justification for dogmattically asserting life arose from purely natural processes? There isn't any evidence to support the claim. What gives?

---

I think the case is really the other way round - we cannot empirically "examine" any other route. We can speculate as philosophers on other Ways in which life could come to be, and many a philosopher has done so. But we cannot investigate other possibilities with the tools of science.I will not speak for others, but I do not dogmatically claim any version of the origin of life. What I do dogmatically claim is that as soon as something is part of the natural world, it is within the sphere of scientific investigation.

quote:

---

What is really sad is that you won’t take your discussion to the International Society for Complexity, Information and Design discussion board.

---

I am registered on the board, read it daily, and hope to contribute some "Brainstorms" on the flaws in Dembski's approach to genetic algorithms soon. I am currently speaking to my employer about permission to discuss some computer science issues online in this context. I do find the ISCID board very interesting indeed, but so far I haven't found anything very persuasive! I don't think I'll fork out the $45 dollars until I'm convinced it's worth it. But I hope one day you will see some Pamboli posts there.

quote:

---

Originally posted by John Paul:
I just posted the positive evidence for ID (a couple posts up) and according to its proponents it can be falsified. They are listed in an article by Dembski to Eugenie Scott.

​

Is Intelligent Design Testable?
A Response to Eugenie Scott

---

Dembski tackles this initially with a subtle play on words

quote:

---

it is a hallmark of science that any of its claims be subject to revision or refutation on the basis of new evidence or further theoretical insight. If this is what one means by testability, then design is certainly testable. Indeed, it was in this sense that Darwin tested William Paley’s account of design and found it wanting. It simply won’t wash to say that design isn’t testable and then in the same breath say that Darwin tested design and refuted it.(Dembski)

---

The implication here is that Darwin in some way scientifically refuted Paley with new evidence or theoretical insight. Darwin, however, did not present evidence against design — he presented a theory which explained the available evidence of variety in the natural world better. Did Darwin therefore provide further theoretical insight? I guess you could say that, but in reality Paley’s argument from design didn’t fail as a theory but as reasoning: it did not stand the test of logic, never mind science.
I don’t know if Dembski is referring to a specific author when he says It simply won’t wash to say that design isn’t testable and then in the same breath say that Darwin tested design and refuted it. In a sense he is right, not because this is a contradiction, but because it is wrong to say that Darwin tested design in a scientific manner.Demsbki later gets on to the meat of his article: that ID is testable in ways that Darwinism is not. He deals first with falsifiability.

quote:

---

If it could be shown that biological systems like the bacterial flagellum that are wonderfully complex, elegant, and integrated could have been formed by a gradual Darwinian process then intelligent design would be falsified on the general grounds that one doesn’t invoke intelligent causes when purely natural causes will do. On the other hand, falsifying Darwinism seems effectively impossible. To do so one must show that no conceivable Darwinian pathway could have led to a given biological structure.

---

This is curious. Darwinism is open to a simple, straightforward and devastating falsification — observing a designer in action. In practice, all we would need is direct, observable, testable evidence of a designer with means, motive and opportunity (as they say in the old detective novels.)In truth, ID and Darwinism can play the same game: retreating into ignorance. Darwinism can always defend itself by saying that developmental pathways have not yet been discovered, ID similarly can say that whenever a pathway is discovered that the object in question was not irreducibly complex and then look around for another one.What I find curious is that ID theorists do not explicitly explore this area — indeed they emphasis that the nature of the designer is not important, even discounting whether the designer is embodied or unembodied. It is as if they have reason to think that the designer may be unknowable, which is a puzzling attitude for a scientist. If Demsbki, et al, have identified an effect, one would expect a serious scientific explorer to look next for the nature of the cause. In all other cases, especially those quoted by Dembski, where we infer design we are also primarily interested in the designer. What kind of forensic science is interested in identifying murder, but is uninterested in the murderer? Can you imagine SETI finding a signal outer space and then saying the nature of the sender is irrelevant? Yet this is exactly what Demsbki does again and again — even though evidence of the designer would be the most overwhelming evidence of design!When discussing the confirmability of ID, Demsbki has taken on the old, but still lively, issue of extrapolation from small-scale to large-scale evolution. This is much discussed elsewhere and I don’t think I need go over the ground again here. Dembski does have this say Yes, there is positive evidence for Darwinism, but the strength and relevance of that evidence on behalf of large-scale evolution is very much under dispute, if not within the Darwinian community then certainly outside of it. This, I think, starts as a very fair comment, but again there is a vagueness of words here to suit his purpose. The phrase very much under dispute refers to the Darwinian community — he is careful I think not to say the scientific community. In other words, I think he is trying to sneak in a sense that Darwinism is far more controversial in its context than it really is. Of course, it is a hugely controversial subject within a certain milieu of North Americans, but I think he is trying to cast his net a little wider than he properly can.Next, bizarrely, he cites SETI as positive evidence for intelligent design! His argument — correct me if I’m wrong, please — seems to be: if we came across a broadcast sequence of prime numbers, we would regard that as designed. And this is positive evidence??? What he seems to be doing is saying that the concept of specified complexity is in itself evidence for design! An interesting concept of evidence — the ontological argument applied to scientific evidence. I can imagine design, therefore design exists!It is frankly astonishing to me that someone who normally takes such care to appear scientific and practical in their approach should be so hopelessly weak when discussing the confirmability of his theory.On predictability, I find myself almost completely in agreement with Demsbki! For me, Darwinism and ID both are descriptive theories — theories of how things are, not how they should be. I agree that Darwinism cannot predict or retrodict specific mutations. In a stressed environment we could certainly predict that certain mutations would be beneficial and that others would lead to extinction, but it seems to me no part of natural selection that beneficial mutations must occur — only that if they do occur they will be selected for survival. Thus, I regard Darwinism as an entirely descriptive theory — and none the worse for that. Predictability just isn’t a big issue for me.Demsbki’s final point is trivial; though he dresses it up a bit, as is his wont. It can be summarized as follows: a capable enough designer could design anything; therefore the explanatory power of ID is at least equal to Darwinian processes. However, if something was designed, then ID could explain that and Darwinism couldn’t, therefore ID has a greater explanatory power than Darwinism.I wish I had thought of this when I was a kid. The Mr Nobody did it argument has far greater explanatory power than the my sister did it argument I used so unsuccessfully for all those years!