The Design Explanatory Filter

[Administrator2]

[Request from the Administrator: when letters are used in place of words, please put the meaning of the letters in the text the first time they are used. This is especially important for our friends who do not live in the United States. Thank you.]

EF = Explanatory Filter (or Design Explanatory Filter)

SOP = Standard Operating Procedure

ID = Intelligent Design

DNA = Deoxyribonucleic acid

CSI = Complex Specified Information

IOW = In other words

IC = Irreducible Complexity

JOHN PAUL

I feel this deserves a thread of its own, so let us go for it here.

As I see it (if you want William Dembski’s view you will have to ask for it at the International Society for Complexity, Information & Design- http://www.iscid.org/ ), the design explanatory filter (EF) is a flow chart set-up as a process guide, as Dembski calls it, “standard operating procedure”(pg. 36 of The Design Inference). It requires input from a dependable source and is only as good as the information it has to work with. It’s purpose is so you cover all bases before design is inferred, where design is not known beforehand.

The first we see of the EF is the beginning of chapter 2, on page 36 of The Design Inference, which begins with the following:

Whenever explaining an event, we must choose from three competing modes of explanation. These are regularity, chance, and design. To attribute an event to a regularity is to say that the event will (almost) always happen. To attribute an event to chance is to say that probabilities characterize the occurrence of the event, but are also compatible with some other event happening. To attribute an event to design is to say that it cannot reasonably be referred to either regularity or chance.

The flowchart itself can be seen in figure 2.1 on page 37.

The EF starts with an event, E. It takes E to a decision box and asks “Does E have a High Probability (HP) of occurring?”

To answer that question, the operator would have to have known what “high probability” refers to and the context in which it is being used. Assuming the person knew that, if the answer was Yes, E would be attributed to Regularity. If the answer was No, E would proceed to the next decision box. (IOW is E covered by any Law)

The next question that faces E, is, “Does E have an Intermediate Probability (IP) of occurring?”

Again, the only prerequisites are the knowledge of what IP refers to and the context it is being used. With that knowledge a decision is made. Yes means E is attributed to chance, that is until evidence to the contrary tells us otherwise. No would send E to the third, and as it stands now, final decision box.

If E makes it to the third decision box, it is asked, “Does E have a Small Probability (SP)of occurring and is E specified (sp)[SP/sp]?”

Three prerequisites: What does SP refer to? What does specified refer to? And the context.

I then used the EF to determine whether or not DNA originated via purely natural processes or Intelligent Design:

OK DNA enters box #1. Does DNA have a high probability of originating via purely natural processes? We have never observed DNA do this. That doesn't mean it can't happen but it just doesn't happen regularly. (IOW there is no Law that covers DNA formation) On to box #2.

Does DNA have an intermediate probability of originating via purely natural processes? Again DNA has never been observed originating via purely natural processes. Even if we put all the amino acids, sugars, and other chemicals we know are in living organisms into a flask, DNA does not form. On to box # 3:

Does DNA have a small probability of originating by purely natural processes? For discussion sake we will say Yes. Things of small probability happen all the time. But we have only observed DNA in living organisms and not once arising outside of one. Is DNA specified? The DNA in living organisms is. So unless it is shown that any DNA sequence would give rise to a living organism, or easier yet, that any DNA sequence injected into an egg cell will give rise to life, the DNA in living organisms could be considered to be specified. Thus specified complexity is determined to be representative of DNA. From what we know, we only observe specified complexity arise via design.

All this means is that with our current level of knowledge it is very safe to infer that DNA was at one time the result of Intelligent Design.

Why did I choose the whole DNA sequence of an organism as opposed to selected segments? That is the first logical step when dealing with determining design by nature or design by intelligence. Look at as much of the object under investigation as possible.

IOW, it is important to know what the parts belong to before determining what the part does and / or why it no longer wants to do it. I also know, through observation, tinkering and experimentation, that very similar parts can be used for very dissimilar functions and very dissimilar parts can be used for similar functions. The point being before determining whether or not pieces have been designed and for what, it is best to first determine whether the object that the pieces are part of is intelligently designed.

As I have stated before:

The next issue I have is that “some sequences of DNA” were specified. What Dembski’s filter would try to deduce is if the DNA of a living organism is the product of design or purely natural processes. If we limit ourselves to segments of DNA one could pick out a segment of alleged ‘junk’ DNA/ pseudo-genes and because of our lack of knowledge about the genetic code, use this seemingly random sequence as evidence that DNA couldn’t be designed because that segment apparently doesn’t code for anything. That would miss the point and count on our ignorance for support.

No one said that 100% of the design had to remain 100% functional within our limited understanding of what functionality refers to. Also just because God is perfect does not remain God’s Creation has to be 100% functional, as we understand it and not 100% of God’s Creation has to exhibit design, as ours minds correlate it. Because if you get down to it even the nucleus of an atom exhibits design. If the strong force that keeps the atom's nucleus together were just 5% weaker we would have a periodic table of 1. No life, no debate. If that force were just 2% stronger, massive nuclei, no stable hydrogen, no life, no debate. Designed or just happened?

The EF is satisfied that DNA of living organisms can be inferred to have at one time, been designed. The EF cannot tell us what happened to that design after eons of replication, mixing and environmental pressures.

But understanding the design can be very useful. The EF gets us to the starting point. It could also help us tell if someone/ something has been tinkering with genomes.

William Dembski on his EF:

http://www.arn.org/docs/dembski/wd_explfilter.htm

* * *

The following was more an EF tangent so I moved it from the “Irreducible Complexity” thread:

Helen:
Barbarian, the concept of IC lies NOT in the sequencing, but in the function. DNA 'works' because the cell knows what to do with it. In this sense, I suppose, it could be said that the cell itself can tell the difference between what is intelligently designed and what is random nonsense genetically.

Scott Page:
The function of DNA is dictated by the 'sequencing', so in reality, the sequence of DNA is very important. As such, one should think that a 'filter' that is purported to be able to tell what is 'Designed' and what is not, biologically speaking, should certainly be able to tell whether or not a given sequence of DNA is designed, a random conglomeration of nucleotides, or arose naturally.

John Paul:
Actually I already covered that:
</font><blockquote>quote:</font><hr />

John Paul:
It has become apparent that you don’t understand ID or the purpose of the EF. The EF is satisfied that DNA of living organisms can be inferred to have at one time, been designed. The EF cannot tell us what happened to that design after eons of replication, mixing and environmental pressures.

</font>[/QUOTE]Scott Page:
Then how can it possibly be of any use at all? If it cannot look at extant material – that’s all we have, right? – and tell if it was designed or not, how, EXACTLY, is this EF to be employed?

John Paul:
The EF is just a procedure used for detecting design.

As for detecting ID in today’s organisms, sure. For example the EF would be useful in detecting genetically engineered alterations to organisms. This would be useful in finding out if the seeds you are planting have been genetically altered. Some people have a problem with genetically modified food.
It could also tell us if there are any Dr. Moreaus out there. Maybe someone is intentionally making X-Men/ Women.

Scott Page:
And of most import, how, EXACTLY, is the EF “satisfied that DNA of living organisms can be inferred to have at one time, been designed”?

John Paul:
We take all of our knowledge on DNA, all of knowledge on design, all of our knowledge on information and complexity, and all of knowledge of living organisms, put it together with the EF and it is clear that DNA isn’t the result of purely natural processes.

Again, the EF is just a procedure to guide you through the process of making a design inference.

The way I see it there is two alternatives for DNA (and life): either it originated via purely natural processes or it didn’t. If it can’t be shown that life arose from purely natural processes, given our level of knowledge and the information available, then it is safe to infer the alternative. When we look at that alternative, life did not originate via purely natural processes, we have at least two choices: Intelligent Design or Divine Creation. Seeing that science already has fields and processes that detect design, choosing ID just seems like the first logical step. If we ever have a process in place that allows us to detect the Divine, we can pursue that option at that time.

Scott Page:
That is the question! Yet you say:
“The EF cannot tell us what happened to that design after eons of replication…”

John Paul:
Just so we are clear- the EF doesn’t tell us, it guides us.
Someone set up a strawman that wanted to look at segments of human DNA (as opposed the genome). Why look at only a few pixels when you can view the whole screen? And when looking at the specified complexity exhibited by DNA, based on our knowledge of DNA and life, the design implications should not be ignored.

Scott Page:
So please, with peer-reviewed support, of course, EXPLAIN exactly how the EF determined this by using only material that has been subject to eons of replication, mixing, and environmental pressure.

John Paul:
You have it wrong. The EF doesn’t determine anything. The EF is SOP when determining design. It is the person(s) using the EF that makes the determination.
Hey look, if you want to falsify my conclusion just present the peer-reviewed literature that shows that DNA can originate via purely natural processes. Here is something you might like to read first:

The RNA World

http://www.panspermia.org/rnaworld.htm

Or simpler yet, give us an example of CSI coming from non-CSI.

God Bless,

John Paul

[ April 29, 2002, 11:04 AM: Message edited by: Administrator ]

 

[Administrator2]

NEILUNREAL

The Explanatory Filter (EF) reminds me a lot of an early artificial intelligence concept called the General Problem Solver (GPS, not to be confused with the Global Positioning System). The GPS was based on the hypothesis that many problems could be solved by reducing the distance between the current state and the goal by the method of solving intermediate sub-problems. It turns out that although this method is capable of solving a large set of problems, there is an even larger class of more interesting problems which it cannot solve. (I’m not saying the class of problems handled by the EF is identical to that handled by the GPS, only similar.)

I conjecture that the EF is capable of detecting whether a solution (e.g. a string of DNA) was derived (evolved vs. designed) using a technique similar to GPS. For solutions arrived at using other techniques, the EF will not give reliable results. Many researchers, including John Holland, have presented evidence that changes in a gene pool are influenced by “meta” operations more powerful than simple GPS-style search.

This doesn’t invalidate the concept of intelligent design, but it does mean that the EF is not capable of distinguishing between design and non-design. It merely gives some hint as to whether a fairly simple form of search was used or not used to solve some specific problem.

 

[Administrator2]

JOHN PAUL

NeilUnreal:
I conjecture that the EF is capable of detecting whether a solution (e.g. a string of DNA) was derived (evolved vs. designed) using a technique similar to GPS. For solutions arrived at using other techniques, the EF will not give reliable results. Many researchers, including John Holland, have presented evidence that changes in a gene pool are influenced by “meta” operations more powerful than simple GPS-style search.

John Paul:
As I stated above the design explanatory filter (EF) is just a process guide. It is there to make sure we cover the basics before making our initial inference. After the initial inference we can than employ the EF to scrutinize that decision. Or we can run with that initial inference.

NeilUnreal:
This doesn’t invalidate the concept of intelligent design, but it does mean that the EF is not capable of distinguishing between design and non-design. It merely gives some hint as to whether a fairly simple form of search was used or not used to solve some specific problem.

John Paul:
Of course the EF can’t distinguish between design and non-design. But it does help us make that determination.

The EF isn’t anything magical. In science, as well as in engineering (which is applied science), a process must be followed so that a project can move forward. If the process isn’t followed vital steps can be missed or put out of sequence therefor giving false results or a faulty bridge (computer, system, building, vehicle, etc.). As stated before, the EF is only as good as the people using it and the knowledge they have at their disposal. What do you think archeologists, anthropologists, arson investigators, forensic scientists, et al. do when conducting an investigation? They have a procedure they follow.

If, from using the EF, Design is our conclusion, all that means is it is safe to infer design with our current level of knowledge. If at some later time new knowledge arises that throws that conclusion into doubt we re-examine the event in question.

 

[Administrator2]

NEILUNREAL

I agree with much of what you (John Paul) and others have said about the Explanatory Filter (EF). It is possible to create a filter which will separate whatever you put into it into two unambiguous categories. My question is: what has one epistemologically gained by such a filter? Possibly nothing. Any real knowledge of design must come from outside the filter. In order to claim an empirical, algorithmic filter which discriminates design from non-design, Intelligent Design proponents first have to show that design has some empirical basis. If you say the EF shows us what is designed, there is zero gain in knowledge – all that has been accomplished is to make two dictionary entries (“ID” and “EF”) which point to one another.



“I can’t define design, but I know it when I see it.” – The EF is a way of defining or formalizing this statement as a whole, but it lacks the power to tease apart its components. (Aside from the point that I feel that the EF is technically rather weak in terms of it’s discriminatory power. In my logical opinion it only discriminates whether or not one very limited type of process was involved in producing the evidence in question.)



I have personal reservations about our philosophical ability to come up with a definition of design. Is a totally abstract painting an example of design – if so, what about a natural simulacrum of a human face in a rock cliff? Are pebbles sorted by a river an example of design – if not, what about pebbles sorted by a caddis fly larva? (I don’t mean we can’t distinguish between the two types of sorting, I just question what we’ve really gained by doing the distinguishing.) Questions of design seem to always result in infinite regress: if the caddis fly’s pebbles were designed because they were produced by a caddis fly designed by God, then why aren’t the river pebbles an example of design if the laws of physics were designed by God?



If “design” is defined as “scientific evidence which supports faith,” then I have yet to see a convincing definition which doesn’t depend on the pre-existence of faith. And if faith is there, why the need for proof? Either everything is “designed,” or else nothing is (everything is, I believe).



To the extent that Western science and theology have co-mingled, they have never merged. The “theo-scientific” view has remained one of the universe being Newton’s giant clockwork, set in motion by God, with sporadic quasi-supernatural odd-balls (e.g. living things, especially man) plunked down inside this clockwork. Naturalism and supernaturalism continuously fight for possession of these odd-balls (i.e. fight to determine what things are and are not odd-balls). But this view is itself a filter – a myth, if you will – a way of bridging the gap between stuff and person – and EF is an attempt to formalize this myth on part of ID . And if this formalization fails, who decides what part of the original myth gets changed?



I just happen to think it’s possible to be a Christian, or any other type of person, and have a different mythology. So that’s why I’m wary of attempts to mingle theology with fairly rudimentary scientific rules-of-thumb.