Durston Cont’d

_{Barry Arrington

February 3, 2009

Intelligent Design}

Share: Facebook; Twitter; LinkedIn; Flipboard; Print; Email

Kirk Durston‘s Thoughts on Intelligent Design

In this thread, I would like to lay out my own thinking regarding a method to detect or identify examples of intelligent design. I then would like to unpack my thinking in a slow, meticulous (pedantic perhaps?) way and, if we can get that far, apply it to a few examples, including a protein, and the minimal genome.

Defining ‘Intelligent Design’:

I commonly see the term ‘intelligent design’ used in two ways. An example of the first way is in a magazine headline I saw this morning:

‘Evolution by Intelligent Design’

The above example is similar to the way ‘planning’ is used in, ‘Success through good planning.’

In this sense, we can define Intelligent Design as the ability of a mind to produce an effect that both satisfies a desired function or objective and might not otherwise likely occur. This ability emerges out of what we understand to be intelligence, defined in <a href=”http://en.wikipedia.org/wiki/Intelligence”>Wikipedia</a> as the capacities to reason, to plan, to solve problems, to think abstractly, to comprehend ideas, to use language, and to learn.

The second way I see the term intelligent design used is:

‘That traffic control system is a beautiful example of intelligent design.’

The usage of ‘intelligent design’ in the above sentence is similar to the usage of planning in, ‘That rescue operation was an excellent piece of planning.’

In this second type of usage, we can define intelligent design as an effect that satisfies a function or objective and requires a mind to produce. Other examples of intelligent design are the Sphinx and the Microsoft Vista operating system.

In the first sense, ‘intelligent design’ is an ability and in the second sense, ‘intelligent design’ is an effect, or result of that ability.

With this in mind, the definition of intelligent design that I will be using in this discussion is as follows:

Intelligent Design: 1 the ability of a mind to produce an effect that both satisfies a desired function and might not otherwise occur. 2. an effect that performs a function and that requires a mind to produce.

I realize that there are other definitions out there, some of which I do not at all agree with (e.g., Wiki’s). In general, most of the definitions of intelligent design that I see are actually specific examples, applications or results of intelligent design, rather than the defining essence of intelligent design. Ultimately, what I want to argue is that examples of intelligent design all required a mind to produce. I then want to argue that intelligent design is the most rational explanation for the protein families and the minimal genome. I will pause here in case anyone wishes to raise a question about what I’ve covered thus far. Then I will proceed to the next step.

Comments

Jerry: Eukaryotic protein coding genes are split in multiple exons (sometimes really many of them) separated by (usually longer) introns. The whole protein sequence is reconstituted by splicing out the introns at the level of mRNA (before translation). That has nothing to do with folding. Folding is a property of the whole protein molecule, not of a single exon. For instance, if you take human myoglobin, a very simple protein of 153 aminoacids, the coding sequence in split into 3 exons. When they are joined, after splicing the introns, the resulting mature mRNA is translated, and the protein folds in one single compact fold, the globin fold. So, single exons do not fold, and are not functional elements: the whole protein sequence is the functional element. It is true that alternative splicing can give birth to variant proteins. That process is certainly important, and it has brought to the end of the classical "one gene - one protein" model. But I don't think we understand really how alternative spicing is controlled or regulated. You must consider that introns are probably extremely important for regulation, although we scarcely understand their role. While protein coding genes represent only 1.5% of human genome, introns represent more than 30%. The extreme fragmentation of protein coding genes in eukaryotes remains, as far as I know, a fascinating mystery.gpuccio_{February 9, 2009
February
02
Feb
9
09
2009
01:59 AM
1
01
59
AM
PDT}

Kirk, I have a simple question about proteins in general. And it is an attempt to improve on what may be a little learning on my part. From what I know there are introns which are interspersed into a gene and are removed after splicing. The remaining exons are then translated into a protein. Somewhere I have heard that multiple genes may be made from the same gene by only splicing together some of the exons. Thus, one gene can make several different proteins depending upon the exons used. Is this correct? And if it is correct does each exon have its own folding properties and does the whole protein have different folding properties from its parts. I understand it is larger and the physics maybe somewhat different but the fact that each exon may fold mean that combinations of the exons will also fold and can they be predicted from how the individual exons fold. This may be naive or I may be talking nonsense. The thought hit me that if folding proteins are very rare in general then the fact that the whole may fold may mean that various combinations of the parts may also fold and be useful. If this is not simple and does not fit into what you plan to present then maybe some time in the future you could answer it.jerry_{February 8, 2009
February
02
Feb
8
08
2009
10:24 PM
10
10
24
PM
PDT}

KF, Sick PC? All you need to know are two things: 1) Avira AntiVir (Free) and 2) Malwarebytes Anti-Malware (also free) I had a dead computer a couple weekends back and norton anti-virus stood by and watched while I got infected. The two above completely removed the infections without having to re-install anything or reformat my system.Atom_{February 8, 2009
February
02
Feb
8
08
2009
09:17 AM
9
09
17
AM
PDT}

kairosfocus[63], Yup. And it doesn't really make a difference for the tiny probabilities we talk about here. Good to be in agreement with my ludlumesque friend! SuperPOProf_P.Olofsson_{February 7, 2009
February
02
Feb
7
07
2009
09:15 AM
9
09
15
AM
PDT}

Atom On the way ou tthe door to a techie for a sick PC. Great! Greet her from us all at UD. Gkairosfocus_{February 7, 2009
February
02
Feb
7
07
2009
09:11 AM
9
09
11
AM
PDT}

gpuccio, I agree, CSI deals with "specification" in general (which is just a subset of something else), and Dembski uses compressible strings as his particular subset, hence why I said Dembski's CSI deals with compressible strings. But your general point is correct, CSI is the general theory, FSCI is an extremely useful particular application of it, using "function" as the well-defined and measurable subset. KF, She is good and close to my side always. Today is our one year five month anniversary (a month away from 1.5 years) and we're still seeking G-d's guidance to help our marriage grow and still madly in love. So things are blessed. Now back to Durston...Atom_{February 7, 2009
February
02
Feb
7
07
2009
09:09 AM
9
09
09
AM
PDT}

JT: KD is talking on Van der Waals interatomic forces, bond rigidity [Proline is the classic on that] etc. In the second instance he is speaking of periodic, ordered crystals, not aperiodic info bearing molecules created algorithmically and rolling up the energy hill (ATP is another case in point, due to the rotating turret molecule ATP synthase.) PO/Sup II: Point taken. 1 in X odds is a looser way but it is used too. GEM of TKIkairosfocus_{February 7, 2009
February
02
Feb
7
07
2009
09:06 AM
9
09
06
AM
PDT}

ok, i give up, html is not cooperating today...Prof_P.Olofsson_{February 7, 2009
February
02
Feb
7
07
2009
08:46 AM
8
08
46
AM
PDT}

Sorry about a forgotten semicolon that made the previous post look ugly. 1 in 6 probability of rolling 61 to 5 odds in favor of rolling 6 5 to 1 odds against rolling 6Prof_P.Olofsson_{February 7, 2009
February
02
Feb
7
07
2009
08:46 AM
8
08
46
AM
PDT}

kairosfocus[28], Tecnical point since you brought it up: "odds" and "probability" are different. Odds give the relative size of a probability of an event to that of its complement. For example, the probability to roll 6 with a die is 1/6 so the odds against rolling 6 is (5/6)/(1/6) usually expressed as 5 to 1. The odds in favor of rolling 6 is the reciprocal, 1 to 5. The preposition is usually key: 1 in 6 probability of rolling 6&nbsp1 to 5 odds in favor of rolling 6. Now I read gpuccio's post, and the point became moot as he actually writes probability 10^-250 which is the preferred way, mathematically speaking.Prof_P.Olofsson_{February 7, 2009
February
02
Feb
7
07
2009
08:35 AM
8
08
35
AM
PDT}

..we went on to publish a paper proposing a measure of FSC... In Hazen’s equation, Functional information I(Ex) is defined as I(Ex) = - log2[M(Ex)/N] (1) ...there are only three major areas we have ever observed FSC. One is human languages, the other is human-designed software, and the third is in biopolymers such as DNA and proteins. Something to think about.
Just some very informal musings- It seems ironic to say that a simple law can distinguish life from nonlife but a simple law could not generate life.
Keep in mind, it is physics that determines which amino acid sequences have stable folds, not biology.
But I thought that physics could only produce simple repetitive patterns. So imagine a retarded child and all he can do is make simple repetitive patterns all day. And yet he can also determine which amino acid sequences have stable folds. Sounds like he and nature are idiot savants. Speaking of simple repetitive patterns, I'm thinking of quasars and black holes and galaxies and weather and tornadoes and earthquakes and volcanoes and the rings of Saturn, and rainbows and million upon millions of galaxies and a universe millions of light years in extent. Seems quite an accomplishment for the laws of physics.JT_{February 7, 2009
February
02
Feb
7
07
2009
06:23 AM
6
06
23
AM
PDT}

GP Excellent thoughts. My only caveat: when we state probabilities as like 1 in 2^500, in English we call that the "odds" of. Gkairosfocus_{February 7, 2009
February
02
Feb
7
07
2009
04:51 AM
4
04
51
AM
PDT}

kairosfocus: thank you for the clear and exhaustive contribution. Since we are finally discussing the whole "package" of specified information and its computation, I would like to add a couple of thought which I have already stressed in previous discussions, and which remain very important for me: 1) Of the two properties of CSI (I am using the term here in the more general, inclusive sense, until we agree on nomenclature), it is "specifictaion" which is the real mark of design. So, in FSCI, specification is directly willed and conceived by the conscious intelligent designer, as a form of specific meaning and function to be expressed. Complexity, on the other hand, can or cannot be present at significant levels. So, if the specification (function) can be achieved with low complexity, then the designer will be happy to do so. The designed object, in that case, remains designed, but it is a designed (functional) object of low complexity. But, when high complexity is necessary to ensure function, human designers are surprisingly good at generating it: see for instance the example of language. The high complexity, therefore, is not essential to designed things, but it is essentially to objectively recognize designed things, distinguishing them form possible pseudo-designed things (objects which appear specified but, being of low complexity, could arise by chance in a random system). 2) To examplify as simply as possible what I have said in point 1), I propose again a very simple example: a string of digital information (let's say binary, for simplicity), which corresponds to the digits of pi. Now, let's suppose that we don't know for certain the origin of that string: let's say that it can be read in some physical series of events or objects, and that such a series could have arisen by chance in a random system, or have been designed by someone. Now, we look at the string. And let's suppose we have 3 strings, one of 8 bits, another of 200 bits, and another one of 500 bits. Now, here we have the same specification in all three cases: the bits are apparently random and, as far as I know, not compressible, and yet, if we recognize that they correspond to the first n bits of pi, they specify a very useful and important mathematical object. So, in the right context, they are absolutely functional. What is the search space of the 3 strings? That's easy: 2^8, 2^200, 2^500. And the target space, here, is easy to calculate: it is 1 in all 3 cases, because only one string of that length is correct, given the specification. So, the probability of each string (assuming an uniform distribution in the supposed random system) is 2^-8, 2^-200 and 2^-500. And the complexity is 8, 200, and 500 bits. Now, are those three strings specified? Yes. But can we say that they are designed by an intelligent agent, or is the specification in them a pseudo-specification, not connected to design? Here the complexity helps. It is easy to see that string number one can easily be a random result, given its very low complexity. And I think that everybody would agree that string number 3 must have been designed. What can we say of string number two? Even if, formally, it is very distant from our usual UPB suggested by Dembski, I would definitely consider it designed. And so would probably do most reasonable people. Indeed, I have often stated that the UPB is really an excessive threshold. But what if the complexity is, say, 30 bits? Here many would have doubts. So, that's why we have to agree on some threshold, and the threshold must be appropriate for the physical system which is supposed to have generated the string randomly. So, different systems will need different thresholds. So, I hope this example helps in clarifying the relative roles of specification and complexity, which are often misunderstood.gpuccio_{February 7, 2009
February
02
Feb
7
07
2009
03:56 AM
3
03
56
AM
PDT}

Folks: Most interesting thread! Atom, how's the luminous one? PO-Superman II, how's the sub-orbital flying into Sweden these days? [Did you run into St Nick last Dec 24/5? How does he do that round the world trick in one eventing?] 1] of ornithopters and ID: First, a point of information: ornithopters have been designed, built and flown. (Cf the video!) So,as with Ventner on production of bio-information by intelligent design, we know that intelligent designers can make a bird-like flying machine. Inetelligent designers are the only OBSERVED originators of bird-like flying machines. Now, the project to make a LIVING birdlike flying machine, that's a hard on for ya. (Maybe, someone can try converting a lizard into a flying animal by genetic manipulation?) 2] Of Specifications and complex sequences Plainly, Wm A D's point -- and he used this example --was to specify a narrow target in a large config/state space, but not by painting the rings around the arrow after it hits. Functionality is one way to do that: as KD points out for biopolymers, function is kept by physics, not by bio. So Bio must only search for the prize, it cannot directly generate it. Compressibility and the like, are other ways to get to narrow targets. A third, more general one is: independent, "simply" statable specification of the target zone. (a) FSCI hits as "it works," which is macroscopically recognisable. (b) Compressibility works as the specifying program is shorter than the original sequence and specifies it. [I note that this does not get away form complexity as the program has to be stated in a language, encoded as a signal, expressed by executing machinery, putting you right back up on complexity.] (c)Aesthetically appealing art object works as the object is digitally definable and can be simply described: Ultra-large, representational, 3-D portraits of four US presidents of note, in a group. (d) convenient card hands work [within the constraint that the deck of cards gives a much smaller space] e.g. 13 spades in a standard deck of 52. 3] FSC metrics, islands of function and probability Here KD et al have put up a metric that allows us to recognise that functionality comes in islands and the islands are subsets of a much larger config space. In the case of proteins of a family, the H2N-CHR-COOH building blocks chain on the H2N- and -COOH backbone, hosting the functional groups, R, as a branch. the basic chaining chemistry and the functionality are attributes of different features of the blocks. then too, the sequencing is coded in DNA, which is itself chained independent of the active parts of the GCAT monomers. So, we have good reason to infer to more of less uniform odds for each possible slot. That only some islands exhibit function is then reasonably modelled by taking the Hazen ratio [M(Ex)/N] as a probability metric. But we are not locked up to such, as KD discusses on ground states: cosntraints may shift from uniform distributions [null state as ground state] and metrics can be made to address that [e.g. weighted sums]. But as he observes [2007, p.4]
Physical constraints increase order and change the ground state away from the null state, restricting freedom of selec-tion and reducing functional sequencing possibilities . . . The genetic code, for example, makes the synthesis and use of certain amino acids more proba-ble than others, which could influence the ground state for proteins. However, for proteins, the data indicates that, although amino acids may naturally form a nonran-dom sequence when polymerized in a dilute solution of amino acids [30], actual dipeptide frequencies and single nucleotide frequencies in proteins are closer to random than ordered [31]. For this reason, the ground state for biosequences can be approximated by the null state.
This is of course a very important observation relative to the many (sometimes heated) discussions previously held in and around this blog. 4] the Durston et al FSC metric: Maybe, I can help us a bit on deciphering, per your remarks on marking clear distinctions? [KD, kindly correct if I miss a key point.]
The measure of Functional Sequence Complexity,denoted as Z, is defined as the change in functional uncer-tainty from the ground state H(Xg(ti)) to the functional state H(Xf(ti)), or Z = [delta]H (Xg(ti), Xf(tj)) . . . Eqn 6 [Using Z for zeta] [P. 4]
--> here we imagine a ground [macro-]state which specifies in effect the set of possible sequences, then we jump to an island of function [functional macrostate]. --> We can see protein chains of given length, and we can see those that are of that length and WORK in the required role --> there is a jump in information, here on a per symbol average basis [which is what H measur5es in info theory generally]: from the generic sequence of length X state, tot he in an island of function state, in effect --> Z is naturally in bits [given its components], and since they are functional, we have a measure in Fits. 5] On a per aa basis: PP 4 - 5:
Consider that there are usually only 20 different amino acids possible per site for proteins, Eqn. (6) can be used to calculate a maximum Fit value/protein amino acid site of 4.32 Fits/site [NB: - log2 (20) = 4.32]. We use the formula log (20) - H(Xf) to cal-culate the functional information at a site specified by the variable Xf such that Xf corresponds to the aligned amino acids of each sequence with the same molecular function f [putt he chains in parallel with aa codes laid out in cols by corresponding sites]. The measured FSC for the whole protein is then calcu-lated as the summation of that for all aligned sites. The number of Fits quantifies the degree of algorithmic chal-lenge, in terms of probability, in achieving needed meta-bolic function . . . . A high Fit value for individual sites within a protein indicates sites that require a high degree of functional information. High Fit values may also point to the key structural or binding sites within the overall 3-D structure. Since the functional uncertainty, as defined by Eqn(1) is proportional to the -log of the probability, we can see that the cost of a linear increase in FSC is an exponential decrease in probability.
--> How to roll yer own. 6] Functional state probabilities:
For the current approach, both equi-probability of mono-mer availability/reactivity and independence of selection at each site within the strand can be assumed as a starting point, using the null state as our ground state. For the functional state, however, an a posteriori probability esti-mate based on the given aligned sequence ensemble must be made . . . [.] [A] set of aligned sequences with the same presumed function, is produced by methods such as CLUSTAL, downloaded from Pfam. Since real sequence data is used, the effect of the genetic code on amino acid frequency is already incorporated into the outcome. Let the total number of sequences with the specified function in the set be denoted by M. The data set can be represented by the N-tuple X = (X1, ... XN) where N denotes the aligned sequence length as mentioned earlier. The total number of occurrences, denoted by d, of a specific amino acid "aa" in a given site is computed. An estimate for the probability that the given amino acid will occur in that site Xi, denoted by P(Xi= "aa") is then made by dividing the number of occurrences d by M, or, P(Xi = "aa") = d/M.(7)
More specifically, continuing:
For example, if in a set of 2,134 aligned sequences, we observe that proline occurs 351 times at the third site, then P ("proline") = 351/2,134. Note that P ("proline") is a conditional probability for that site variable on condi-tion of the presumed function f. This is calculated for each amino acid for all sites. The functional uncertainty of the amino acids in a given site is then computed using Eqn. (1) [I.e. H(Xf(t)) = -[SUM] P(Xf(t)) logP(Xf(t)) (1), where Xf denotes the conditional variable of the given sequence data (X) on the described biological function f which is an outcome of the variable (F) . . . , p. 2 ] . . . using the estimated probabilities for each amino acid observed. The Fit value for that site is then obtained by subtracting the functional uncertainty of that site from the null state, in this case using Eqn. (4), log20. The individ-ual Fit values for each site can be tabulated and analyzed . . . [.] The summed total of the fitness values for each site can be used as an estimate for the overall FSC value for the entire protein and compared with other proteins.
--> he goes on to discuss changes in Z on mutations etc . . . 7] Evolving proteins and changing Z: Pp 5 - 6:
In principle, some proteins may change from a non-func-tional state to a functional state gradually as their sequences change . . . . Intuitively, the greater the reduction in FSC a mutation produces, the more likely the mutation is deleterious to the given function. This can be evaluated using known mutations introduced individually into a set of aligned, wild-type sequences to measure the change in FSC. The results could then be ranked. Operating under the hypothesis that mutations producing the greatest decrease in FSC are most likely to be deleterious, experimental investigations into certain genes with certain mutations could be prioritized according to how negatively they affect FSC . . .
of course the now famous table of 35 values follows. WELL DONE, KD! GEM of TKIkairosfocus_{February 7, 2009
February
02
Feb
7
07
2009
03:13 AM
3
03
13
AM
PDT}

Atom (#52): I agree with you, only I would consider CSI as the general set (any kind of specified complex information), and compressible or functionally specified information as two distinct subsets of CSI. But, obviously, it's just a matter of agreeing on terms and definitions. The substance remains the same.gpuccio_{February 7, 2009
February
02
Feb
7
07
2009
12:24 AM
12
12
24
AM
PDT}

JT: I am not stating that animals are not conscious. I believe they are. And as such, they cannot, IMO, be completely explained by current materialistic theories. But what I was stating is that animals cannot generate functional complex information: language, software, machines. I am not denying that animals can have some form of language or some form of intelligence, I am just saying that our definition of FSCI is not matched by what they do. And finally, my statement: "Until differently proven, intelligent agents remain an empirically observable reality, and they behave in a characteristic way, and have properties which cannot be found elsewhere." is in no way an argument from ignorance. I am stating observable things. And I am saying that the "theory" that those observable things (consciousness and intelligence) can be explained on the basis of other observed things (material objects and the inherent material laws) which have different properties and behaviour is just a theory, and needs specific support to be taken in consideration. Where is this an argument from ignorance?gpuccio_{February 7, 2009
February
02
Feb
7
07
2009
12:21 AM
12
12
21
AM
PDT}

Response to ID D: Response to JT: JT asked, ' Why can’t (or shouldn’t) you consider the agent as part of the physical system.' By physical system, I mean a system that is described/prescribed by the laws of physics. The laws of physics, described by simple equations, most of which do not exceed even one line of text, do not have the horsepower to explain intelligence. The onus would be on the person who wants to suggest that physics can explain, say, thinking, to show it or model it. Belief in leprechauns and belief that physics can explain thinking have something in common …. the complete absence of any evidence for either. Of course, if someone believes that human thought can be fully described by the laws of physics, then they would need to test that theory. Mere conjecture does not constitute science. I think the paper on three subsets of complexity, mentioned in my last post, is a propos here. The laws of nature tend to produce effects that are repeatable, thus, have very little capability for producing functional information. re. Birds as a model for aircraft: Birds themselves are an anomaly in the physical system. In the old days, God or the gods were credited with creating them. Now we have the tools to test at least the hypothesis that there were created by intelligent agency, through measuring their functional complexity. Of course, if an intelligent agent is successful in building an artificial bird, that is an exercise in intelligent design. In the same way, any successes we have in designing a new protein, or building an artificial life form, are also examples of intelligent design. Re. compressibility and patterns: Patterns and compressibility do not distinguish between functional and non-functional information (meaningful information and gibberish). The laws of physics can produce repeating patterns (ordered complexity), such as in a crystal lattice, as well as sequences that so far as we can see cannot be compressed (random complexity), such as atactic polystyrene. For biological life, it matters a great deal whether a sequence is functional (functional complexity), not whether the sequence is compressible or contains patterns. Rob Only a knowledge of the laws of physics is required to determine what constitutes the ground state.KD_{February 6, 2009
February
02
Feb
6
06
2009
05:25 PM
5
05
25
PM
PDT}

JT wrote:
But everything in his arguments starts with the observation regarding the small percentage of strings that are compressible. Then later he says at one point,...
Both FSC and CSI deal with small subsets: FSC (as expounded by Durston et al.) deals with the small subset of functional states among total possible states; CSI (as expounded by Dembski) deals with small subset of compressible strings among all possible strings. Both deal with subsets and sets, just different kinds. Durston's is easier to deal with (IMHO) and is more focused and concrete, hence why we usually use FSCI around here, rather than the more generic CSI. Just my two cents. AtomAtom_{February 6, 2009
February
02
Feb
6
06
2009
05:20 PM
5
05
20
PM
PDT}

Everybody, Could we perhaps abstain from very long comments? We want Kirk to be able to finish before this thread also gets long and starts loading slowly. Rememeber why it was started in the first place.Prof_P.Olofsson_{February 6, 2009
February
02
Feb
6
06
2009
03:48 PM
3
03
48
PM
PDT}

gpuccio wrote [45,46]:
On that other thread I already made the point that functional biological information is not particularly compressible, if not at all. What is there of compressible in the sequence of a functional protein? therefore, functional information is perhaps a subset of the more general concept of CSI, but not a subset of “compressible” information.
We could go back and dissect Dembski's writings to settle this I suppose, but my characterization of them is based on a conscientious and thorough study of them undertaken within the limits of my abilities only within the last several weeks (You may recall an extended discussion on them here not too long ago.) But everything in his arguments starts with the observation regarding the small percentage of strings that are compressible. Then later he says at one point, (and paraphrasing now, but an accurate characterization nonetheless): "The key to ruling out chance is to keep the pattern simple." In fact, the measure of CSI in the actual formula is inversely proportional to the complexity of the pattern detected. There may be something regarding the "detachability" of a pattern that eludes me and is perhaps not strictly tied to algorthmic compressibility, I don't know. (Or maybe I'm not missing anything.) But the bottom line is, all that can be ruled out by this method is randomness. There is nothing in the whole procedure that can tell you, "This was caused by 'design' as conceived in I.D. circles and not by laws or mechanism." I don't think ruling out randomness is worthless though, as I do believe that evolution does largely equate to randomness unless most of the information came from the natural laws as opposed to the mutations.
Regarding your other points, you make the usual standard objections of materialists, who try to deny the empirical nature of consciousness. I would remark that “intelligence” is a property of a conscious agent, and that nobody has ever demonstrated (notwithstanding all the arrogant statements of strong AI theory) that consciousness and intelligence can be explained as a consequence of the objective and mechanical laws of physics. Consciousness remains an empirical observation, unexplained by current materialistic theories. That is what is called “the hard problem of consciousness”. Therefore, you cannot out of dogmatic authority reduce intelligent agency to a product of mechanical laws. Until differently proven, intelligent agents remain an empirically observable reality, and they behave in a characteristic way, and have properties which cannot be found elsewhere. Intelligence, and the ability to output functional information easily and with very high complexity, are among them.
I'm merely repeating some pretty commonsensical observations here, but to me, animals seem "conscious" but I for one am not going to assign mystical transcendent attributes to animals. An animal's behavior, its being, its internal life, is attributable to its physical-chemical makeup. It seems to me a sort of hubris stemming from misconceived religious dogma, to just assume that humans operate according to some entirely different set of mystical principles. I think your argument above is obviously an argument from ignorance, as you say, 'Until proven otherwise, I will assume such and such..." That's the definition of an argument from ignorance. My philosophical stance OTOH would be one of practicality - If human behavior is the result of something akin to I.D.'s conception of intelligence, then it is not potentially decipherable in the way that the rest of nature is. I would say that whatever we could potentially understand about intelligence is of neccesity quantifiable, measurable, observable, physical. And to the extent intelligence is not these things, then it is meaningless.
[JT:] “Wouldn’t you suppose that the reason that human-designed software exhibits FSC is because the “biopolymers such as DNA and proteins” that resulted in human beings exhibit FSC as well?” [gpuucio:]I would never suppose that. Plants and lower animals are very rich in FSC in their DNA and proteins, but I am not aware that they design software. Am i missing something in your assumption?
The things that a human's internal organs do exhibit a lot of genius - the function of the liver the heart, and so on, presumably all the FSC in them is directly attributable to FSC in human DNA. Of course, higher animals share all these attributes with us. But animals can perform remarkable feats of dexterity as well, that if they are broken down and analyzed are incredibly complex. I'm thinking of the dexterity of a cat, for example. Certainly many animals manifest various types of genius that we could never hope to duplicate. Mankind has his own type of unique genius as well, undoubtedly.JT_{February 6, 2009
February
02
Feb
6
06
2009
03:02 PM
3
03
02
PM
PDT}

gpuccio[48], I don't disagree with you. I don't know if JD does either. I merely tried to explain what I think JD said when he asked "wouldn't you suppose...." in which case your objection regarding plants and lower animals would not be valid. I was going to make up a metaphor about Italy and soccer, but you get my point!Prof_P.Olofsson_{February 6, 2009
February
02
Feb
6
06
2009
02:57 PM
2
02
57
PM
PDT}

Prof_P.Olofsson: But we have no evidence that possessing FSC in the DNA is a necessary, even if not sufficient, condition. As far as we know, the only condition necessary to design is to be conscious intelligent agents, as is proved by the observed (both subjectively and objectively) connection between the process of design and specific conscious representations, intentions, desires, and so on. And, unless we have solved the hard problem of consciousness (and we haven't) nobody can affirm that possessing FSC in the DNA is a necessary condition to produce consciousness. So, let's stay with what we know for certain: being a conscious intelligent agent is the only "necessary" empirical condition connected to the process of design, and to the production of designed objects. That's an observable fact. As we don't know what makes humans conscious intelligent agents, all the rest is assumptions.gpuccio_{February 6, 2009
February
02
Feb
6
06
2009
02:29 PM
2
02
29
PM
PDT}

gpuccio[46], I think you might be turning JT's logic around. I think he is saying essentially that you suppose "only FSC can produce FSC" not that "any FSC can produce FSC." In logical terms, possessing FSC in the DNA etc is a necessary condition but not a sufficient condition to produce FSC. Or something like it. Sorry JT if I misinterpret you.Prof_P.Olofsson_{February 6, 2009
February
02
Feb
6
06
2009
02:01 PM
2
02
01
PM
PDT}

JT: "Wouldn’t you suppose that the reason that human-designed software exhibits FSC is because the “biopolymers such as DNA and proteins” that resulted in human beings exhibit FSC as well?" I would never suppose that. Plants and lower animals are very rich in FSC in their DNA and proteins, but I am not aware that they design software. Am i missing something in your assumption?gpuccio_{February 6, 2009
February
02
Feb
6
06
2009
01:34 PM
1
01
34
PM
PDT}

JT: On that other thread I already made the point that functional biological information is not particularly compressible, if not at all. What is there of compressible in the sequence of a functional protein? therefore, functional information is perhaps a subset of the more general concept of CSI, but not a subset of "compressible" information. Regarding your other points, you make the usual standard objections of materialists, who try to deny the empirical nature of consciousness. I would remark that "intelligence" is a property of a conscious agent, and that nobody has ever demonstrated (notwithstanding all the arrogant statements of strong AI theory) that consciousness and intelligence can be explained as a consequence of the objective and mechanical laws of physics. Consciousness remains an empirical observation, unexplained by current materialistic theories. That is what is called "the hard problem of consciousness". Therefore, you cannot out of dogmatic authority reduce intelligent agency to a product of mechanical laws. Until differently proven, intelligent agents remain an empirically observable reality, and they behave in a characteristic way, and have properties which cannot be found elsewhere. Intelligence, and the ability to output functional information easily and with very high complexity, are among them.gpuccio_{February 6, 2009
February
02
Feb
6
06
2009
01:29 PM
1
01
29
PM
PDT}

[42] correction: relative = relevant, Para. 4.JT_{February 6, 2009
February
02
Feb
6
06
2009
11:38 AM
11
11
38
AM
PDT}

KD:
The functional complexity of a system is the change in functional uncertainty (defined in the paper) between the ground state and the functional state.
That's where it seems that things might get tricky. Sorry to bring up the diamond example again, but in regards to carbon configurations, you proposed three different ground states. These ground states had N=2, N=1, and N=large number, respectively. Presumably, our identification of the ground state is based on the conditions under which the observed configuration was formed. This means that the accuracy of our FSC calculation depends on our knowledge of the configuration's causal history. Is this correct? If so, I'm sure you can guess what my next point will be.R0b_{February 6, 2009
February
02
Feb
6
06
2009
11:28 AM
11
11
28
AM
PDT}

KD [17,22,39]: Just some feedback directed only to the comments you've made in this thread:
There is one type of circumstance, however, where all intelligent agents must exercise their ability for intelligent design. It occurs when the agent desires some function or objective and the physical system is not likely to cooperate by producing it. The agent then exercises that ability and what results is not only an example of intelligent design (definition 2), but an anomaly within the physical system. It is anomalous in virtue of the fact that the physical system was not likely to produce it.
Why can't (or shouldn't) you consider the agent as part of the physical system.
For example, an observer gazing at the skies between AD 1000 and the present, would have seen an empty sky save for the usual clouds, birds, etc. However, early in the 20th century, the observer would have seen something anomalous, an aircraft. People desired a function or held an objective to fly. Nature did not seem very helpful in satisfying this objective, so humans exercised their ability for intelligent design (def. 1) and produced a piece of intelligent design (def 2) that would fulfill that function. The resulting aircraft in the sky was an anomaly within the physical system.
I notice you mentioned birds as part of the preexisting default background noise. When an intelligent agent would have come up with his own flying machine it would have been as a result of hours and hours and hours of observation of birds, and then there would be an attempt at duplication of something that already existed in nature.
By ‘required’ I do not mean that it would be nomologically or logically impossible for nature to produce the effect. I merely mean that the probability that nature could produce the effect becomes so low within the boundary conditions of the problem that the intelligent agent must exercise intelligent design (def 1) and it would be irrational to believe otherwise.
On another thread today I brought up compressibility in the original concept of CSI, wherein if some sequence of sufficient length indicates any kind of pattern whatsoever this rules out chance (as the percentage of compressible strings is exceedingly small according to Dembski). I was admonished that functionally specified information is the only relative concept within the context of biology. It is this concept that I believe you have attempted to formalize and are alluding to in the above paragraph, as something that exceeds the probability for nature to produce. Well my point would be that compressiblity is already an extremely exclusive set. Any sequence of sufficient length exhibiting it I do believe we could rule out as happening by metaphysical chance. So what point does it serve to focus in on a narrower subset of compressibility (functionally specified information) when for the purposes of the probability argument, compressibility will suffice. I presume its because you would have to say the entire universe is designed, since we see patterns everywhere. But focussing on functionally complex specified information I don't believe solves your problem. You couldn't say mechanism ("laws") didn't produce it, only that randomness did not. For the record I do personally think that there is of necessity a direct correlation between the compexity of nature and the complexity of man. Furthermore I think that any set of laws and preexisting conditions that resulted in us would effectively equate to us, i.e. "Man" in a different form, in a different, prexisting phase.
At this point, I will propose an hypothesis as follows: H: a unique property of intelligence is the ability to produce significant levels of functional information. Note that I have claimed that this property is ‘unique’ to intelligent agents.
What if as it turns out that humans do what they do solely by virtue of the configuration of their physical attributes. Then intelligence would be merely a property of a physical system. Would it be sufficient for you that any system producing life be labelled "intelligent"? If evolutionists as a concession one day started describing the mechanism they propose (such as it is) as "intelligent", would that satisfy I.D.? Is there anything in your arguments that establishes that intelligence or mind is something other than natural or physical?
As Abel and Trevors point out in their paper on three subsets of sequence complexity, there are only three major areas we have ever observed FSC. One is human languages, the other is human-designed software, and the third is in biopolymers such as DNA and proteins. Something to think about
Wouldn't you suppose that the reason that human-designed software exhibits FSC is because the "biopolymers such as DNA and proteins" that resulted in human beings exhibit FSC as well?JT_{February 6, 2009
February
02
Feb
6
06
2009
11:25 AM
11
11
25
AM
PDT}

Prof_P.Olofsson: Just a question to you, to start the discussion. I see that Hazen's equation: I(Ex) = - log2[M(Ex)/N] is assuming, if I am not wrong, an uniform probability distribution. As you know, I comletely agree with that position, but I would like your comments about that. And this is not Dembski, or any other ID source. This is a paper on PNAS. And it is exactly the same kind of computation that I have suggested many times, in discussions with you and others, both here and on Mark's blog. So, can we discuss here this problem of distributions in a deeper and more objective way?gpuccio_{February 6, 2009
February
02
Feb
6
06
2009
10:42 AM
10
10
42
AM
PDT}

Kirk: thank you for the wonderful post. First of all I have to say that I have always admired your paper, and quoted it a lot of times, both here at UD and on another blog, as the only example of easy and immediate computation of functional complexity in proteins. So, I am very happy that we are able to discuss it with you directly here. I think that Abel and Trevors have given a very clear theoretical foundation to the concept of fucntional information in biology, but I didn't know that the practical application of Shannon's H to that computation was your personal ides. My most heartfelt compliments for that! I have many things that I would like to say about the computation of the target space in proteins. It is a complex and fascinating issue (and a very fundamental one). And it is one issue often used by darwinists to raise obscure objections. I can easily predict that we will see some of them here very soon. So, I would rather wait for the discussion to develop, and then offer some personal thought if it is necessary.gpuccio_{February 6, 2009
February
02
Feb
6
06
2009
10:26 AM
10
10
26
AM
PDT}

Prev 1 2 3 4 Next

You must be logged in to post a comment.

Leave a Reply