There is a lot of debate in the comments to recent posts about whether the argument from ID is circular. I thought it would be worth calling this out as a separate item.
I plead that participants in this discussion (whether they comment here or on UD):
- make a real effort to stick to Lizzie’s principles (and her personal example) of respect for opposing viewpoints and politeness
- confine the discussion to this specific point (there is plenty of opportunity to discuss other points elsewhere and there is the sandbox)
What follows has been covered a thousand times. I simple repeat it in as rigorous a manner as I can to provide a basis for the ensuing discussion (if any!)
First, a couple of definitions.
A) For the purposes this discussion I will use “natural” to mean “has no element of design”. I do not mean to imply anything about materialism versus supernatural or such like. It is just an abbreviation for “not-designed”.
B) X is a “good explanation” for Y if and only if:
i) We have good reason to suppose X exists
ii) The probability of Y given X is reasonably high (say 0.1 or higher). There may of course be better explanations for Y where the probability is even higher.
Note that X may include design or be natural.
As I understand it, a common form of the ID argument is:
1) Identify some characteristic of outcomes such as CSI, FSCI or dFSCI. I will use dFSCI as an example in what follows but the point applies equally to the others.
2) Note that in all cases where an outcome has dFSCI, and a good explanation of the outcome is known, then the good explanation includes design and there is no good natural explanation.
3) Conclude there is a strong empirical relationship between dFSCI and design.
4) Note that living things include many examples of dFSCI.
5) Infer that there is a very strong case that living things are also designed.
This argument can be attacked from many angles but I want to concentrate on the circularity issue. The key point being that it is part of the definition of dFSCI (and the other measures) that there is no good natural explanation.
It follows that if a good natural explanation is identified then that outcome no longer has dFSCI. So it is true by definition that all outcomes with dFSCI fall into two categories:
- A good explanation has been identified and it is design
- No good explanation has yet been identified
Note that it was not necessary to do any empirical observation to prove this. It must always be the case from the definition of dFSCI that whenever a good explanation is identified it includes design.
I appreciate that as it stands this argument does not do justice to the ID position. If dFSCI was simply a synonym for “no good natural explanation” then the case for circularity would be obviously true. But is incorporates other features (as do its cousins CSI and FSCI). So for example dFSCI incorporates attributes such as digital, functional and not compressible – while CSI (in its most recent definition) includes the attribute compressible. So if we describe any of the measures as a set of features {F} plus the condition that if a good natural explanation is discovered then measure no longer applies – then it is possible to recast the ID argument this way:
“For all outcomes where {F} is observed then when a good explanation is identified it turns out to be designed and there is no good natural explanation. Many aspects of life have {F}. Therefore, there is good reason to suppose that design will be a good explanation and there will be no good natural explanation.”
The problem here is that while CSI, FSCI and dFSCI all agree on the “no good natural explanation” clause they differ widely on {F}. For Dembski’s CSI {F} is essentially equivalent to compressible (he refers to it as “simple” but defines “simple” mathematically in terms of easily compressible). While for FSCI {F} includes “has a function” and in some descriptions “not compressible”. dFSCI adds the additional property of being digital to FSCI.
By themselves both compressible and non-compressible phenomena clearly can have both natural and designed explanations. The structure of a crystal is highly compressible. CSI has no other relevant property and the case for circularity seems to be made at this point. But FSCI and dFSCI add the condition of being functional which perhaps makes all the difference. However, the word “functional” also introduces a risk of circularity. “Functional” usually means “has a purpose” which implies a purpose which implies a mind. In archaeology an artefact is functional if it can be seen to fulfil some past person’s purpose – even if that purpose is artistic. So if something has the attribute of being functional it follows by definition that a mind was involved. This means that by definition it is extremely likely, if not certain, that it was designed (of course, it is possible that it may have a good natural explanation and by coincidence also happen to fulfil someone’s purpose). To declare something to be functional is to declare it is engaged with a purpose and a mind – no empirical research is required to establish that a mind is involved with a functional thing in this sense.
But there remains a way of trying to steer FSCI and dFSCI away from circularity. When the term FSCI is applied to living things it appears a rather different meaning of “functional” is being used. There is no mind whose purpose is being fulfilled. It simply means the object (protein, gene or whatever) has a role in keeping the organism alive. Much as greenhouse gasses have a role in keeping the earth’s surface temperature at around 30 degrees. In this case of course “functional” does not imply the involvement of a mind. But then there are plenty of examples of functional phenomena in this sense which have good natural explanations.
The argument to circularity is more complicated than it may appear and deserves careful analysis rather than vitriol – but if studied in detail it is compelling.
I think I am a bit to blame for allowing standards to fall. But you can moderate your own thread Mark. Feel free to send this to guano as an example!
The word functional is nearly a synonym for selectable in the context of evolution, so the only issue to be decided is whether the history of genome sequences is confined to neutral or positive sequences.
Today is a good day to begin this discussion because Lenski’s paper is now available online.
http://bms.ucsf.edu/sites/ucsf-bms.ixm.ca/files/20121011.strauli.nicolas.pdf
Any discussion of what can be done by natural processes must henceforth be informed by what has been observed in the confines of a few beakers over little more than a decade. Of particular interest is the necessity of non-functional potentiating mutations. The relevance to gpuccio’s thesis is that these necessary precursors cannot be intelligently selected because their importance only becomes evident in hindsight.
The Lenski experiment reaches or exceeds Behe’s Edge in just years and validates many hypotheses regarding the role of duplication and exaptation.
I’ll contribute this, from a comment of mine in the other thread. It’s based on Dembski’s argument as presented in Specification: The Pattern That Signifies Intelligence.
re:
This sounds suspiciously like Sherlock’s “whatever remains, however improbable, must be the truth”.
Yet the problem is I just don’t see how it can be done in any practical sense. Unless you have a complete knowledge of unguided evolution or any other unintelligent process then you simply can’t say.
I believe this is also the reason why KF, Gpuccio defend their spurious “probability” based arguments so strongly.
If they can claim it’s “needle in a cosmos” unlikely then what need to actually obtain perfect knowledge of unguided evolution or any other unintelligent process?
Only an unreasonable person would ask for that, no? So once that leap has been taken everything else follows naturally.
OMTWO,
Yes, and Dembski actually includes that quote in his paper:
OMTWO:
Dembski tries to gloss over this problem:
The irony of course is that there are none of those anyway. I’m yet to see anything approaching an “explanation” from an ID perspective.
Sad thing is that Joe has it nailed, “it was designed is an explanation” (aka design is a mechanism) is all that can be said but only he actually says it.
Mung
You write:
and later:
I am only using natural to mean not–designed. If you prefer to substitute the word material or any other word feel free to do so. It should not affect the argument provided it is understood that it means not designed.
Alan I am sure you are less guilty than most. It is well-known phenomenon of internet discussion and we all fall into the trap. Certainly I do. (Lizzie is the only exception I know). Just thought it would do no harm to issue a bit of a call to arms.
Mung:
I’m open to the possibility of an objective indicator of design. It’s just that I haven’t seen one yet that isn’t circular and/or unreliable. CSI and dFSCI are circular. Irreducible complexity is not circular, but it’s not a reliable indicator of design.
Circularity is a feature of the logical structure of an argument. We don’t have to agree with your premise to identify circularity.
A maxim uttered by a fictional character – and put in his mouth by a man who believed in fairies!
“Eliminative” induction sounds like BS to me.
I know in Medical diagnosis they use the phrase “rule out” a lot. But that’s a Bayesian kind of induction, where you have a list of possible diagnoses and you are trying to find the most likely.
ID proponents are being irrational when they list a never observed agent as having a significant probability of being the cause of something.
It’s exactly the same as asking the police to give equal weight to ghosts as perpetrators of crimes.
But even in eliminative mode, it is wrong to rule out evolution, particularly now that we have established examples of evolution captured in action. What is the phrase? Hyperskepticism?
Mung,
Imagine that you are a caveman. You are walking one day and you walking into the middle of New York circa now.
Empty of people.
Do you suppose that the caveman would attribute different things to design then you would from not having your lifetime of unconscious learning about what is and what is not designed?
Would the caveman walk through New York and out the other side, perhaps wondering at the variety of nature?
http://3.bp.blogspot.com/_BaxMlnoKZyY/SXVxU9eAkOI/AAAAAAAABew/orv560yV-Bs/s400/mamallapuram-rock-india.jpg
http://withfriendship.com/images/a/1451/Mahabalipuram-Balancing-Rock.jpg
http://i.ytimg.com/vi/PvjgZBiybUg/0.jpg
Mark writes, in the OP:
Mark,
The addition of qualifiers doesn’t (and can’t) save these concepts from circularity. It’s a simple matter of logic, and it doesn’t depend on what the qualifiers are.
Suppose that I want to prove that some object has property D. I define a characteristic — let’s call it SuperDuperness — and I claim that it is a surefire indicator that an object has property D.
Here’s how it works: If an object has properties A, B, C, and D, then it is SuperDuper. If it lacks any of these properties, then it is not SuperDuper. I claim that 100% of SuperDuper objects turn out to have property D. SuperDuperness is a 100% reliable indicator. And I’m right, but only because my argument is circular. An object has to have property D in order to be declared SuperDuper, so by definition all SuperDuper objects have property D.
It doesn’t matter what properties A, B, and C are. The fact that property D is a requirement for SuperDuperness is, by itself, enough to doom the argument to circularity, and the only way to break the circularity is to get rid of the dependency on property D.
Mung:
It’s simple, Mung. Here’s how I explained the circularity on the other thread.:
If gpuccio eventually becomes convinced that it could have evolved, you have a false positive.
No, then he just concludes that it never should have been considered to have dFCSI.
I congratulate Mark, keiths, and others for clearly pointing out the circularity issue. I (and before me, Shallit and Elsberry) had arguments against the Law of Conservation of Complex Specified Information, which was used to show CSI could not get into the genome by natural selection. But then I didn’t quite know what to do about the P(T|H) step in the CSI argument.
I think now it is clear that Dembski is not longer defending the LCCSI but is invoking (as gpuccio does) the P(T|H) term to rule out natural selection as the agent. Pointing out that this renders the whole CSI diagnosis irrelevant is important. It makes clear that the diagnosis of irreducible complexity is the important step in ruling out P(T|H), and that further argument beyond that is irrelevant. So Dembski’s Design Inference argument has in effect been dropped in favor of Behe’s arguments. (There is also the NFL stuff, which has been effectively been rebutted by many people, and the Search For a Search argument, which does not argue against the effectiveness of natural selection).
I am reminded of a previous discussion (I can’t remember where but it may have been Mark’s blog) in which a calculation of CSI for proteins was being talked about. n^20 etc. I suggested to the commenter (computerist) that they might just as well pick any arbitrary (see, UB!) number for the upper probability bound. He seemed to say it wouldn’t matter as everything is designed, anyway.
This would appear to be an obstacle to mutual understanding. Whether things are designed and micromanaged on an ongoing basis by some sort of untraceable imaginary entity seems a matter of conviction and the mathematics serves only to obscure rather than clarify.
Would anyone agree that “circular” is synonymous with “default” in the sense that “therefore design” is the final line in the argument, or as others have pointed out, a “Sherlock Holmes” argument (made by an imaginary character invented by a man who believed in fairies, indeed 🙂 )?
Aside to mung:
To establish someone is lying, you have to show that what they are saying is wrong and also that they know what they are saying is wrong when they say it. You would perhaps find people might begin to take you seriously if you were a little more restrained in using the “liar” one-liner accusation. Feel free to ignore this as you have before.
625
MungOctober 19, 2012 at 5:36 pm
Mung, you might consider you have a rôle to play too in moving dialogue forward. But if you think you have responded with grace and substance then I am a little doubtful of your powers of self-reflection!
Mung <a href=”http://www.uncommondescent.com/intelligent-design/id-foundations/the-tsz-and-jerad-thread-continued/#comment-437157″>writes</a>
<blockquote>I want to see Mark explain how your definition of dFSCI can result in a false positive and yet at the same time be circular.
If it’s circular, what is the false positive?
They have to deny that there can be any false positive.</blockquote>
Don’t know about anyone else, mung, but I don’t think we need to worry about false positives in imaginary design detection yet. It has yet to be established what “design” might entail and how an instance of “design” might be distinguished from anything else. Can there be such a thing as a true positive test for design? The explanatory filter as been abandoned (default argument <i>nec plus ultra</i> and we are left with Dembski’s CSI and Behe’s irreducible complexity again both with default assumption of “design”.
Gpuccio
Thank you for your response. It must have taken a lot of time and effort (11 pages of A4). The good thing is that it enables me to rephrase my argument using your vocabulary. I hope this will by-pass a lot of irrelevant disagreements and get to the essence of things. As I see it there are two parts to the discussion of cicularity.
Circularity due to the “necessity” clause
I disagree with you about the nature of explanations but I am quite happy for the purposes of this argument to replace “good explanation” with “known origin” which seems to be your preferred way of putting it.
You write that for something to have dFSCI it must:
“have no known necessity mechanism that can generate the string”
and later
“no credible necessity explanation”
I take it these mean the same thing and I refer to this as the “necessity clause” in dFSCI (CSI has a similar clause but I won’t get into an argument as to whether it is the same).
As I understand the necessity clause is part of the definition of dFSCI. If a necessity mechanism is found that can generate the string or greatly increase the probability of that string arising then that string does not have dFSCI by definition? Can I confirm that?
Now the key thing is that I cannot see the difference between a “necessity mechanism that can generate the string” and a “non-designed origin”. To me they are the same thing. An origin is something that either creates the string or contributes to the creation of the string and this increases the probability of that string occurring. And that is why it seems to me dFSCI must have a designed origin by definition. Maybe I have misunderstood what you mean by necessity mechanism. In that case can you explain the difference illustrated with examples?
Circularity due to Function
Again we could have a long dispute over the meaning of the word “function” but I don’t think it is necessary. I will work with your definition of “function” as a role that the observer imposes on the object. Given this definition then I agree it is not circular. On the other hand, given this definition, many things are functional, specified, and complex and have non-design origins. Consider the function of making the Sun’s corona visible during the day. The moon is exactly the right size for doing this during a total eclipse. What is the probability of it having that diameter given all the possible diameters it might have had? Almost zero. So using your definition of “function” the argument from design is not true by definition. It is false.
Mung
I apologise that I don’t have time to respond to all your comments. I will pick up on 624. The example of fire and smoke is a good one and rather helpful.
For brevity I define “non-fire” as something which does not involve fire.
To me the ID argument is analogous to this:
1) Define “smoke” as “a phenomenom which for which there is no known non-fire cause/explanation/origin (Gpuccio prefers origin you seem to prefer explanation – I am not particular)”.
2) In all cases of “smoke” where the cause/explanation/origin has been determined it turns out to be fire. (Here is the circularity).
3) Here we have a another phenomenom for which the non-fire cause/explanation/origin is not known. We cannot identify a reasonable non-fire cause/explanation/origin. Therefore, it is “smoke” and from (2) above we have empirical evidence it was fire.
Of course real smoke has other properties – colour, consistency etc. And it is not circular to investigate whether all phenomena with that colour and consistency are caused by fire. And indeed we would find that in many cases they are not caused by fire. Similarly FSCI has other properties as well as “no reasonable non-fire cause/explanation/origin” which is why I went on to investigate them. As I tried to explain above, once I had understood Gpuccio’s definition of “function”, it is quite clear that these properties often have a non-fire cause/explanation/origin which does not include design.
Gpuccio
I just noticed your 657:
I count eight questions in that post. I provide my answers below but the questions seem beside the point. I don’t think either Keiths or I dispute that Shakespeare’s sonnet was designed. The question is why? And can that argument be extended to life. But out of politeness, here are my answers:
“Why is my verse so barren of new pride,”
etc., in a system where coins are tossed, and the output transformed into letters? Do you believe that? No.
Mark to gpuccio:
Mark,
Could you explain why you think that adding the “function” requirement to the definition of dFSCI removes the circularity? I think it doesn’t (see this comment), and I am curious to know where you think my reasoning goes astray.
Gpuccio
I am short of time right now. So I will answer your shorter comment 660 and leave the longer one until tomorrow.
a) Could you explain how do you think we infer that the sonnet is designed, and the other string is not?
Bayesian inference. I can think of many different hypotheses which include an element of design which (a) I know are very likely to exist (b) are far more likely to produce the sonnet than any hypotheses that are without an element of design and are likely to exist.
b) Would you agree that the dFSCI property as defined by me can be well assessed for both strings, according to your same answers, as present in the sonnet, and absent in the other string?
I am not sure because I am quite confused about your view of the difference between origins, mechanisms and explanations. Tis distinction (which as far as I can see is unique to you) is quite new to me. The answer to this will have to wait until tomorrow.
c) Do you agree that such an assesment is in line with what we independently know of those two strings’origin?
This clearly depends on the answer to (b).
Gpuccio
OK I a little time to respond to your your comment 658. I still believe the circularity is there – but this distinction between the mechanism and the origin complicates the argument. Understanding your definition of dFSCI leads to a slight alternation in my position – but I think it has no practical importance. I believe that your definition of dFSCI means that by definition if something has dFSCI then either design plays a role in generating the outcome or it depends on a bizarre and incredible coincidence. Therefore there is no way that a plausible non-design account can lead to an outcome with dFSCI and this follows from your definition – there is no need to observe anything.
The easiest way to explain this is to consider the example you give:
If such a random number generator generated the sonnet just once we might just regard it as the most bizarre coincidence but we would certainly see it as one off. This is not relevant to explaining the features of life that have dFSCI where it certainly requires more than one bizarre coincidence. However, if a psaeudo random number generator could generate the sonnet with any frequency then it is clearly not a very good random number generator. That follows from what we mean by “random” in this context. In your terminology it is a necessity mechanism. We may not understand how it works – but if it can produce the result with any reasonable frequency then it is a necessity mechanism and therefore, by definition, the sonnet no longer has dFSCI. I very much doubt that in the case of the sonnet such a mechanism exists but if such a mechanism does exist then there is no dFSCI. Therefore, there it is not surprising that all cases of dFSCI have an element of design.
I also want to say some more about the “digital” and “functional” parts of dFSCI – but I am in a great hurry again.
Your definition would avoid the circularity is you eliminated #4. Then it would just be wrong, rather than circular.
We’ve already shown that your original definition doesn’t encapsulate your actual demarcation. Rather, #4 should read “no uncontrovertible non-design cause is known”, which eliminates strongly supported scientific findings that evolution can produce functional complexity.
Controvertible is an odd word to use here. It means disputable or opposable by reason. Most scientists would consider evolution so well-established that it is not reasonably disputable as a scientific matter. Even granting that all scientific conclusions are considered tentative, it would be “perverse to withhold provisional assent”. That means #4 would not exclude evolution. However, you would exclude evolution. At the very least, it means #4 is ambiguous. More so, it is the very thing you are trying to prove in #5.
It’s #5 that makes your argument circular, pivoting on your personal opinion. It should read,
#5 Any object whose origin is not be due to non-design causes according to gpuccio is designed according to gpuccio (without exception).
Of course it is without exception.
Keiths – I am sorry I owe you a reply. I don’t really think that adding the other attributes of dFSCI such as digital, specified and functional removes the circularity resulting from the “no necessity mechanism” condition. However, I am trying as hard as I can to see things from the ID perspective and I think they might see the digital, specified, functional attributes as being the essence of dFSCI and the “no necessity mechanism” as as often present not definitive.
It is like defining smoke as small particles in the air usually produced by fire as opposed to defining it as small particles in the air produced by fire.
Some IDers may hold that position but gpuccio has repeatedly maintained that dFSCI is only present if no “necessity mechanism” is known to be responsible. Here’s one of his more recent statements to that effect:
That’s pretty explicit.
Gpuccio 693
I fear we have not made as much progress as you think.
It is useful that you accept that by definition if something has dFSCI then either design plays a role in generating the outcome or it depends on a bizarre and incredible coincidence. But I cannot see what role empirical testing can play. Rather than a lengthy discussion of your specific reasons for making observations I will issue a simple challenge which I think summarises my case.
Describe to me a observation, other than one which we know in advance to be so improbable it would never happen, that would falsify the statement “everything that exhibits dFSCI is designed”.
I bet you can’t do it. And if you can’t then what is the value of an observation when you know in advance what the result is going to be?
His problem is the step where he excludes RV or RMNS. He does so on the basis of lots of information, the very thing he is trying to demonstrate can’t result from RMNS. Among other problems, he hasn’t demonstrated how many steps it takes to get from a random sequence to an optimized sequence.
So he has no empirical basis for quantifying information. By any rational standard, “digital information” is that which is selectable or which potentiates a later selectable mutation. The raw length of a sequence is not particularly important.
Onlooker is right. Gpuccio has repeatedly stressed the “no necessity mechanism” (henceforth “NNM”) criterion for dFSCI. Zachriel has even suggested to gpuccio (at least twice) that he remove the criterion, thereby eliminating the circularity, but gpuccio won’t do it.
It’s not hard to see why. Without NNM, dFSCI means nothing more than “dFSI greater than the threshold”, which in turn means nothing more than “couldn’t be produced by pure RV without selection”. It’s a “tornado in a junkyard” argument, and it’s irrelevant to the evolution debate because it ignores selection, which is an indispensable part of Darwinian evolution. Gpuccio understands this and knows that dFSCI thus defined would be stillborn. Hence his refusal to abandon NNM.
The problem, of course, is that adding NNM dooms dFSCI to circularity:
Gpuccio: I say that X couldn’t have evolved.
Questioner: Why do you say that?
Gpuccio: Because I’ve determined that X exhibits dFSCI.
Questioner: How do you know that X exhibits dFSCI?
Gpuccio: It has dFSI greater than the threshold, and I don’t see how it could have evolved.
Questioner: So in other words you don’t think it could have evolved because it exhibits dFSCI, and it exhibits dFSCI because you don’t think it could have evolved. Sounds circular to me.
Gpuccio: It’s not circular!
Gpuccio
You wrote:
But I am not asking to give me a real example. I am asking you to give me an example of a hypothetical observation which could reasonably happen in any universe you care to describe which might falsify the statement:
“everything that exhibits dFSCI is designed”
(where “reasonably” means with some acceptable probability)
This is nothing to do with what empirically does exist. It is to do with what conceptually could exist. Your empirical convictions are irrelevant.
Your point about the Moon example is a good one. What happened with the Moon example is that I dropped the “no necessity mechanism clause” without thinking about it. Clearly the Moon has the diameter it does because of some natural process. It would for example be physically impossible for a body that was much larger to orbit the earth. I actually did this subconsciously because I was focussing on the functional side of FSCI. The Moon does not exhibit FSCI according to your definition.
It does however exhibit FCSI according to another closely related but different definition. This one says something has FCSI if it has a functional specification and if the probability of it meeting that specification is astronomically low based on a uniform probability distribution (There are in practice many different uniform probability distributions that could be selected for real outcomes but there is generally one that the ID theorist finds obvious such as all base pairs are equally likely and are independent of each other). In the case of the moon the assumed probability distribution is something like all possible diameters from 0 to a very large but not defined maximum. According to this definition other possible mechanisms and their associated probability distributions are not considered. This definition of FSCI is not circular. But as the Moon example shows, it is not always associated with design. You can only make it always true by defining FSCI as excluding cases where there is a possible mechanism.
Gpuccio rests his case against evolution on a couple of premises. One is the isolated islands of function assumption. Another is that sequences having no obvious cousins have no common ancestors. The danger of the latter assumption is that it fails in observable situations.
One obvious example where it fails is in tracing the history of human languages, most of which can be linked by similarities and differences that fit a geographical and historical context. In some cases we have written fossils of the intermediate languages.
Except for Basque, which sits in the middle of Europe without cousins.
Mung
Overnight you have made a quite extraordinary number of comments and many of them contain many questions. I don’t know where you find the time. I apologise if I miss some directed to me. I couldn’t come close to reading them all.
However, I pick up on this which I think is useful:
No. Because I can easily imagine an event that did not conform to the laws of gravity. If some objects floated off into space or heavenly bodies did not follow elliptical orbits for example. All I am asking for is a similar hypothetical description of an event which would falsify the statement “everything with dFSCI is designed” using Gpuccio’ definition of dFSCI.
You haven’t made a statement about red things which is a candidate for being circular! I am not saying Gpuccio’s definition of dFSCI is circular. I am saying the statement “everything with dFSCI is designed” is circular.
It may help if I give an analogy. To me the statement:
“everything with dFSCI is designed”
is a bit like saying
“all bachelors are unmarried”
that is circular because it is true by definition. You don’t need to go round asking bachelors about their marital status to prove it. One way you can demonstrate this (if it were not obvious) would be to ask someone to describe a married bachelor. Of course they can’t.
Mung 737
By a circular statement I mean true by definition – a tautology if you like. It is just rather more subtley hidden than in the case of unmarried bachelors.
Gpuccio (not me) has defined dFSCI as including the property “no necessity mechanism”. This is analagous to bachelors being defined as having the property “not married”. After some exploration of what Gpuccio meant by “necessity mechanism” it turns out by his definition that there are only three ways that something could be arise:
1) Design
2) Via a necessity mechanism
3) Something which was so improbable it effectively could never happen (we don’t discover it could never happen – we define it as something which is so improbable it could never happen)
So if dFSCI is defined as having the property did not arise from (2) then by definition is must have arisen from (1) or (3). And by definition (3) never happens so by definition dFSCI must always be designed.
If you doubt this logic the easiest way to disprove it is to give me a hypothetical example of something with dFSCI that was not designed and it not defined to be so improbable it would never happen- much as I gave the example of planets without elliptical orbits to show the law of gravity is not true by definition.
Mung
I cannot work you whether you are serious or just playing a game!
You write:
But I didn’t write that it is a circular definition. I wrote that it is circular because it is true by definition. Maybe it would help if I dropped the word “circular” and just said that
“Anything with dFSCI is either designed or the result of a bizarre conincidence that could never happen” is true by definition.
I am using Gpuccio’s definition of dFSCI which we have hammered out over many posts. I don’t plan to repeat it.
Mung
I am not on vacation and getting rather behind so this will probably be my last comment on the subject. You write:
I don’t know what I wrote that gave you the impression that I thought the definition of dFSCI is circular. It is no more circular than the definition of “bachelor”. Whatever it was I am sorry, it must have been unclear (it would be interesting to know what it was).
All I am saying is that the statement “things with dFSCI are designed (or so improbable they could never happen)” is true by definition. It may be a good idea to drop the word “circular” altogether. I am still waiting for a counterexample to disprove this claim. All you have to do is use your imagination to think of a hypothetical example of something which is not designed and is not defined in such a way that it could never happen and which results in dFSCI.
Can someone parse that so it makes sense?
Um, doesn’t “uncontrovertible” mean precisely not controvertible?
Functional complexity eliminates randomization. Then #4 eliminates known “deterministic explanations”. Per your posited trichotomy, you’ve eliminated everything other than design.
Of course, evolution is not a random cause, nor is it a deterministic explanation. So you haven’t eliminated evolution. If you group it within random or deterministic, then all you’ve done is added window dressing to the question of evidence for evolution.
Gpuccio,
I think I see the source of your confusion on the circularity issue. You keep stressing the word ’empirical’. For example, you write:
You appear to think that if there is an empirical observation somewhere in the argument, then the argument can’t be circular. That’s simply not true. Circularity is a logical characteristic of an argument, so an argument can be circular regardless of the empirical status of any of the statements within it.
For example, suppose I define “borky” to mean “weighs more than ten pounds.” I then say: “It’s an empirical fact that borky things weigh more than ten pounds.” That’s true. It’s even empirically true. But the argument is still circular. Using borkiness as an indicator of an object’s weight doesn’t tell us anything we didn’t already know. We had to determine that the object weighed more than ten pounds before we classified it as borky.
Likewise, you say that something exhibits dFSCI if 1) it has dFSI greater than the threshold, and 2) you are unaware of any “necessity mechanism” that could have produced it. Suppose you then note that “it’s an empirical fact that nothing that exhibits dFSCI, when it’s true origin is known, was produced by a necessity mechanism”. That’s true. It’s even empirically true. But the argument is still circular. Using dFSCI to indicate that something wasn’t produced by a known ‘necessity mechanism’ doesn’t tell us anything we didn’t already know. We had to determine that that known necessity mechanisms couldn’t produce it before we classified it as exhibiting dFSCI.
What makes gpuccio’s argument circular is that he asserts that a chain of events is impossibly improbable if the chain exceeds a certain number of elements, and he counts only the elements that are part of an impossibly long chain.
His definition of improbable is self-referential. He doesn’t see this because he thinks he has independent evidence for impossibility.
If you found a bank account with a million dollars in it, and you were told it was all compound interest from an initial investment of a dollar that was found on the street, You would judge the possibility of the story being true by the amount of time involved and the history of interest rates.
In other words, you would judge the merits of the story based on facts about the characteristics of the investment environment. You could not simply say that a million dollars exceeds some limit.
I’m sure gpuccio bases his argument on the work of Douglas Axe, rather than on raw numbers. In other words, he is arguing that the interest rates don’t support the compound interest story.
Axe’s argument is non-circular, but it is irrelevant to dFSCI. It would not take 80 bit leaps to invalidate evolution. Behe is essentially correct that requiring leaps of three or four co-dependent mutations would render evolution unlikely if the mutations taken separately involve a selectable loss of function.
Which is why research needs to continue along the lines of Lenski and Thornton. We need a number of data points detailing the functional landscape and the resources available to traverse it. What Lenski and Thornton have done is demonstrate that the landscape is not so sparse that sideways exploration is impossible. The water between islands is simply not that deep.
This is why I have said from the first encounter with gpuccio that his concept is vacuous without knowing the actual history of sequences. dFSCI really only applies to gaps — discontinuities. Sequences without extant cousins are not discontinuities. They are simply bits of the present for which we do not know the history.
Beautiful!
This cuts the Gordian knot of the circularity issue. As I understand gpuccio’s position, there is no situation in which he would ever respond with (A). Since (B) is the only possible answer, it must be obvious even to him and his supporters that his “empirical conclusions” are in fact simply a consequence of his definition.
Nicely done.
Where did the comment I was responding to go?
Gpuccio
(modified from my first attempt after rereading your comment)
Oh dear we go round and round in circles. Let’s clear up one thing (which still confuses Mung). I am not saying that our definition of dFSCI is circular. The word “circular” seems to be causing all sorts of confusion so I work with this way of stating it:
The statement “things with dFSCI are designed (or so improbable they could never happen)” is true by definition.
(I would like to be a bit more precise about the phrase “so improbably they could never happen” because of course, as Mung says, vastly improbable things happen all the time – but it would be time consuming and start off yet another line of misunderstandings and definitions).
To me this statement is synonymous with
The statement “things with dFSCI are designed (or so improbable they could never happen)” is circular.
However, I think you may mean something else by “circular” so let’s not waste time on semantics.
I see you mention the possibility
You say the statement is not true by definition but empirically true. I think you are deluding yourself. You think it is empirically true but actually if you analyse it carefully it isn’t. Just as someone might think it was empirically true that for all circles was the circumference is just over three times the radius (You can imagine them making repeated observations and believing that the evidence was building up!).
The test is – if it is empirically true then it must be logically possible (although empirically false) that something with dFSCI was created by a process which was neither designed nor highly improbable. All you have to do to prove your case is describe such an process. What is it that you might have observed but didn’t? I have been asking for this example for about two days now and you have been unable to produce it.
You mention a hypothetical process X which might generate dFSCI which is not design or unbelievable coincidence – but you don’t describe it. In the end this comes down to the term “necessity mechanism”. I don’t believe there is any logical space for a process that is not design, or unbelievable coincidence, or necessity mechanism. Your definition rules out necessity mechanism and logically leaves you with only design and unbelievable coincidence. If there is a logical space, the process X, all you have to do is describe it!
Here is my understanding of why there is a problem with gpuccio’s argument. We are investigating a DNA sequence.
1. We conclude that it has some relevant “function”.
2. We conclude that it is extremely improbable that it could have that much function (or more), even once in the history of the universe. Thus it constitutes Complex Specified Information, and its high degree of function could not plausibly have resulted from a pure chance mechanism such as random mutation.
3. We check whether deterministic mechanisms such as natural selection could have brought about this high degree of function. We conclude that they could not have done this. (I leave aside the issue of whether natural selection is properly described as deterministic).
4. Now we are ready to see whether the sequence has dFCSI.
But wait! We now already know the answer — the sequence must have been designed. There is no need to assess dFCSI, beyond steps 1-3.
Another way of describing the problem with dFCSI is that it is redundant to steps 1-3. Once they are accomplished there is no further need for the concept of dFCSI. Arguments that step 3 has been accomplished are typically Michael-Behe-style arguments. dFCSI does not add anything essential to them.
The above argument would be incorrect if gpuccio has some way of assessing the presence of dFCSI that does not require step 3.
Sorry onlooker I deleted my comment moments after I posted it and replaced it with one below (which oddly did not include the example you liked so much). As you liked the example I will repeat it here as best as I can remember.
It was addressed to Gpuccio.
You believe that some proteins have the property of dFSCI. Imagine that some scientists discovered how these proteins developed from simple material sources through material process without any unbelievable coincidences (I believe this is already true – but imagine they did it to your satisfaction). Would you then say:
A) Here is a refutation of the statement that everything with dFSCI is designed
or
(B) This shows that those proteins didn’t have dFSCI after all.
That’s another great way to put it.
Mark,
Thank you for returning to patch the temporal disturbance I nearly created.
We now have practically the entire population of TSZ trying to explain to gpuccio why his argument is circular.
Any lurkers want to register and try their hand at presenting an argument, metaphor or analogy that will finally get through to gpuccio? (I’m still hoping that my “borkiness” analogy will succeed, but I’m also realistic.)
Borkiness is an excellent analogy.
The only way to be more clear would be to replace the term “dFSCI” with the actual definition in one of gpuccio’s claims. For instance, he recently said “Things that exhibit dFSCI will never be empirically generated by a random system or by a necessity system, or by a random system helped by necessity effects.” (He followed this with “That is empirically true by definition.” which is nonsensical any way I can come up with to parse it.)
If we replace “dFSCI” with gpuccio’s stated definition, we get “Things that exhibit [high functional information (excluding RV as an explanation) and have no known necessity mechanism that can explain them] will never be empirically generated by a random system or by a necessity system, or by a random system helped by necessity effects.” That’s a clear tautology.
Unfortunately, I am banned from any threads started by kairosfocus for the unforgivable crime of pointing out his hypocrisy and intellectual cowardice, with examples.
I think that onlookers can clearly see the IDists getting emotional which calls their statements into question.
Why would someone with a better position get emotional?
Clearly, they believe they’re losing the scientific argument.
As long as they have their Joes’s and Mungs’s, we will win in front of school boards.
So Joe and Mung, keep up the bad work! 🙂