Design Operates at Multiple Levels

_{Barry Arrington

February 19, 2010

Intelligent Design}

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

In a comment to a prior post lastyearon writes:

I’m simply not understanding how it is possible to detect that certain things were the result of design if everything is the result of design. If you hold that the laws of nature were Fine-Tuned for life, then that position seems incompatible with the notion that it is possible to detect that certain things were the product of Intelligent Design. IDers say they can detect design by distinguishing designed objects from products of natural ‘undirected’ causes. But if natural causes were designed for life, then doesn’t that invalidate that claim?

I reply:
You seem to imply that “IDers” are the only ones who claim to be able to distinguish between designed objects and objects that are the result of undirected causes. This is simply untrue. Here are two strings of text:

String 1: Uq[49epfia[epfoias[efojafpojuawer89yup9fj0075v9aus[-er uqpw\\dflkjoigjeriodfdfioaergoierioadf;lkdfrgerkiergsdfvm

String 2: To be, or not to be–that is the question: Whether ’tis nobler in the mind to suffer The slings and arrows of outrageous fortune Or to take arms against a sea of troubles And by opposing end them.

Now you tell me. Which one of these strings of text is designed and which one is random gibberish. I am certain you will answer that it is obvious that string 2 is designed and string 1 is random gibberish, and so it is. There, you’ve detected design. And anyone else would reach the same conclusion whether they are an IDer or an inveterate opponent of ID theory.

To answer your question, consider this analogy. If I go to Home Depot I will see piles of lumber, nails, paint, wire, etc., in short, everything I need to build a house. No one believes those materials found their way into the aisles of Home Depot by natural undirected processes. They were manufactured and delivered by intelligent agents. But still there is no house. Thus we see that the materials at Home Depot are necessary, but they are not sufficient to build a house. The house will only ever be built if an intelligent agent assembles the materials in a complex and specified fashion.

In the same way, the ID proponent says that the finely tuned laws of nature are necessary for the existence of life, but they are not sufficient. What is missing? Complex specified information. And the fundamental premise of ID theory is that complex specified information arises ONLY as the result of the acts of intelligent agents. So you see, just as in the Home Depot example, design operates at two levels. It operates at the level of setting the conditions (building materials ready to be used; finely tuned laws of nature), and it also operates at the wholly separate level of the design of specific things (building the house; building the DNA molecule).

Comments

Which, as usual, brings us around once again to the origin of the genetic code. The ID argument for this is essentially that the probability of generating the code from scratch is so small as to be non-existent. However, given the foregoing analysis, this is equivalent to saying that the probability of any single set of lottery numbers winning tomorrow's lottery is so small as to be equally non-existent. Yet, someone will win the lottery, if not tomorrow then in the following drawing (or short series of drawings). Furthermore, the probability that this set of numbers being a winner – 1 2 3 4 5 6 – is exactly the same as this set of numbers being a winner – 4 8 15 16 23 42 (but don't buy a ticket with the second string, or you will have to share the prize money with a very large number of people if it wins). The common calculation of the a priori probability of any particular genetic code (i.e. vanishingly small) is fatally undermined by at least three phenomena: 1) there may be some necessary relationship between the code and the things it specifies (i.e. a relationship that is the outcome of an as-yet-undiscovered natural law) 2) the mechanisms by which the code originated may be something like the method with which I generated the second string in comment #2, rather than the method by which I generated the first string 3) none of the analyses presented to date have included the fact that the genetic code is meaningful information (i.e. neither Shannon nor Kolmogorov information). This last point is perhaps the most important, as the meaning inherent in the genetic code (i.e. the necessary relationship between the sequences of nucleotide bases, amino acids, proteins, and phenotypic traits) is basis for all of biology. Unless one includes the meaning of the genetic code (and Drs. Dembski and Marks do not), any calculations of the probability of biological evolution are quite literally meaningless (pun intended...inevitable, even).Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
07:42 AM
7
07
42
AM
PDT}

BTW, Allen, Your coin flipping is biased. Remind me to call tails if you ever flip the coin for a football game. :)Paul Giem_{February 20, 2010
February
02
Feb
20
20
2010
07:39 AM
7
07
39
AM
PDT}

Heinrich (#20), You are looking at the wrong problem. I was able to identify the "complex" specified string rather rapidly and securely (a p value of <0.0000001 is pretty significant). Furthermore, the identification had predictive value. If we had more numbers, the next four would have been 1101. The claim you are making is that design detection is subjective. But everything we do is subjective. That's why philosophers sometimes substitute inter-subjective for objective. But inter-subjectivity sometimes becomes universal enough so that we begin to suspect (IMO rightly) that we are dealing with an objective phenomenon. (Remember, if you disagree, you have no basis to declare that ID, or even YEC, is science. After all, it's their subjective experience, and who are you to challenge?) You, like Allen MacNeill did before you, are still missing the point. The point is not that we can detect all design. The point is that we sometimes can. I just did it. Or do you disagree with Allen on this?Paul Giem_{February 20, 2010
February
02
Feb
20
20
2010
07:36 AM
7
07
36
AM
PDT}

And I forgot to close the blockquote following the citation to Wikipedia.Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
07:27 AM
7
07
27
AM
PDT}

Sorry, the last line of the next-to-last paragraph in comment #21 should read 2exp1,000, not 4exp1,000. My bad.Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
07:26 AM
7
07
26
AM
PDT}

Thank you, Heinrich, for that was indeed one of the points I was trying to make with my presentation of the two strings. Another one of the points I was trying to make is that the first string is immensely more "complex" than the second string. That this is the case is easily demonstrated by the fact that exactly duplicating the second string is possible using a very simple algorithm. Indeed, all one has to specify is the starting number (i.e. "0") and the ending number (i.e. "1100") and the method of counting. By contrast, exactly duplicating the first string using the method originally used to generate it is for all intents and purposes (pun intended, of course) impossible. So, as to Paul Glem's guess that the second string is "designed", the answer is...well, not "yes", but "sort of". It's the result of a pattern generator, but that's all. It's actually a remarkably "simple" string, easily replicated exactly using a very simple algorithm. However, and more interestingly, Paul Glem's intuition that the first string is non-random is definitely incorrect. He based his intuition on the "overabundance" of zeros in the string. This is the same phenomenon that causes people to generate strings of ones and zeros that are detectably pseudorandom. When asked to do generate such a string, people tend to put almost exactly as many ones as zeros regardless of the length of the string, whereas in a genuinely random string (such as the one I generated for comment #2), the string is what it is, and what appears to be an overabundance of zeros is an artifact of the shortness of the string. Which leads us to Heinrich's comment that the detection of CSI is absolutely dependent on the observer's prior knowledge. Once I (and Paul) tell you what the algorithm for generating/interpreting the second string is, you can see it immediately. However, even knowing the rules of simple probability led Paul (and probably many of the people reading these comments) to misjudge the "quality" of the first string. Let me assure you, it really is genuinely random because of the way I generated it, and therefore the probability that I could exactly recreate it is so small as to be essentially zero. Now consider the number of bits in the string of nucleotide base pairs in a relatively simply living organism. For example, uropathogenic E. coli (the bacterium commonly associated with urinary tract infections) has about 5,231,428 base pairs. the probability of specifying this sequences of base pairs using the same method I did for generating the first string is 1/4exp5,231,428 (that is, the number of different nucleotide bases multiplied times itself 5,231,428 times). This number is immensely larger (indeed, incomprehensibly larger) than the so-called "universal probability bound", which Dr. Dembski has recently calculated as 1 in 10exp150. This would seem to very strongly imply that constructing and operating an E. coli bacterium is impossible, yet E. coli do it about every 20 minutes (under optimal conditions). How do they do it? The answer was first suggested by Watson and Crick in April of 1953:
"It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material." [A structure for deoxyribose nucleic acid. Watson, J.D. and Crick, F.H.C. (1953) Nature, 171, pp. 737-738]
In brief, the pairing mechanism for the DNA base pairs makes what would otherwise be impossible absurdly easy, by making the specification of the sequences of bases on one strand of the DNA a necessary outcome of the sequences of the bases on the complementary strand. This is analogous in some ways to "specifying" the outcome of flipping a coin 1,000 times:
"[I]f a coin is tossed randomly 1000 times, the probability of any particular outcome is roughly one in 10exp300. For any particular specific outcome of the coin-tossing process, the a priori probability [for] this pattern [occurring] is thus one in 10exp300, which is astronomically smaller than Dembski's universal probability bound of one in 10exp150. [modified from http://en.wikipedia.org/wiki/Universal_probability_bound ] However, if one were to line up a string of 1,000 quarters and then specify that the sequence of a complementary string of quarters is simply head-with-tail, then again the outcome of the pairing process would render the probability of the second (i.e. complementary) sequence as 1, not 4exp1,000. THE POINT: What appear at first glance to be purely random sequences of bits (e.g. the strings of ones and zeros in comment #2) may not be random at all, or may indeed be completely random. However, there is no a priori way to distinguish between these two circumstances. Yes, one can distinguish between them a posteriori, but that's no better than "predicting" the winner of yesterday's lottery drawing after reading the winning number in today's paper.
Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
07:21 AM
7
07
21
AM
PDT}

Joseph - that may have been part of Allen's point. If calculation of CSI, and hence detection of design, relies on the observer's background knowledge, then design detection is subjective. The problem in this case is the specification: how do you a priori specify the patterns that you are looking for. It's impossible to go through every possible pattern (this is related to the underdetermination problem in the philosophy of science).Heinrich_{February 20, 2010
February
02
Feb
20
20
2010
06:14 AM
6
06
14
AM
PDT}

Allen MacNeill:
How about these two strings: 01000110111000000000101000010000011010 and 01101110010111011110001001101010111100 Is there any detectable design or pattern in either one? To be specific (pun intended), which of these strings exhibits “complex specified information” (”CSI”)? Does one string exhibit more CSI than the other? How do you know? And what does this tell us about our ability to detect design in very simple strings of bits (such as are found in a DNA molecule)?
Just so you know science is not conducted in a vacuum. There is always more evidence to consider- other than the object/ string in question. We don't study the object in isolation. So your parlor games may be amusing bvut they really miss the point.Joseph_{February 20, 2010
February
02
Feb
20
20
2010
05:44 AM
5
05
44
AM
PDT}

CJYman and Kairosfocus have been shedding light on practical and detailed organized-CSI calculations for the past few weeks now, the best I've seen.computerist_{February 20, 2010
February
02
Feb
20
20
2010
05:00 AM
5
05
00
AM
PDT}

Mung in #13:
"You’ve admitted that the strings you presented are simple."
Nope. I've only stated that the second string 01101110010111011110001001101010111100 can be recreated by a relatively "simple" algorithm, i.e. counting upwards from zero in binary code. Let me ask it as clearly as I can: Is the first string "complex" (i.e. not "simple") and how do you know?Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
04:56 AM
4
04
56
AM
PDT}

Mung, go back and read all of my comments, above. Where, in any of them, did I either state or imply that either string was "simple"? In fact, one of the two strings is anything but "simple". Again, I ask, which one?Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
04:52 AM
4
04
52
AM
PDT}

Allen @11:
And (finally) re Mung in #7:
Finally? lol
Does the demonstrated fact that you were completely unable to detect even the simplest possible pattern in the second string, despite Paul Glem’s ability to detect it virtually instantly, give you any reason to question the concept of “complex specified information” or our ability to detect it?
The demonstrated fact? Demonstrated by whom? ...the demonstrated fact that I was completely unable to detect even the simplest possible pattern? The simplest possible pattern?
Or, asked another way, is there any empirical/mathematical method by which you could have discovered and verified how the two strings were generated before I told you how I did it?
Was either of your strings complex? If it's not a complex string, it's no wonder that no complex specified information can be found in the string. How much complexity can be embedded within the simplest possible pattern?Mung_{February 20, 2010
February
02
Feb
20
20
2010
04:22 AM
4
04
22
AM
PDT}

Allen @10:
Also in Mung’s comment #7: “…you’ve already assured us that DNA is just simple strings.” Really? And exactly where have I done this?
Allen @2:
And what does this tell us about our ability to detect design in very simple strings of bits (such as are found in a DNA molecule)?
So there are both simple and complex strings of bits found in a DNA molecule? So why don't you provide an example of a complex string? How do you distinguish between simple and complex string?
BTW, Mung, I did give you a “complex specified” string in my original example. Which one was it?
I don't think you did. 1. You indicated the strings were simple.
...what does this tell us about our ability to detect design in very simple strings of bits
2. How did you determine which one of the strings is complex and shich one is simple?Mung_{February 20, 2010
February
02
Feb
20
20
2010
04:04 AM
4
04
04
AM
PDT}

Allen_MacNeill:
Mung in comment #7: Which of the two strings is more “complex”? You have clearly indicated that neither string is “complex”.
No Allen, what I clearly indicated what that you had failed to establish that either string was complex, as clearly indicated by your claim that both strings were "simple." Are you denying that you claimed that these strings were simple? Are you asserting that you claimed that these strings were complex? Allen, we've interacted before. You should know by now that your credentials mean nothing to me. What matters is your ability to state and defend an argument. You've admitted that the strings you presented are simple. Are you now denying your original statement? Are you now claiming that these strings are complex?
You have clearly indicated that neither string is “complex”.
Why would I possibly think that? And why would you possibly think that an ID advocate would think otherwise? My clear claim is that you are demonstrating an unacceptable ignorance of the ID position.Mung_{February 20, 2010
February
02
Feb
20
20
2010
03:43 AM
3
03
43
AM
PDT}

BTW, Mung, I did give you a "complex specified" string in my original example. Which one was it?Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
03:35 AM
3
03
35
AM
PDT}

And (finally) re Mung in #7: Does the demonstrated fact that you were completely unable to detect even the simplest possible pattern in the second string, despite Paul Glem's ability to detect it virtually instantly, give you any reason to question the concept of "complex specified information" or our ability to detect it? Or, asked another way, is there any empirical/mathematical method by which you could have discovered and verified how the two strings were generated before I told you how I did it?Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
03:33 AM
3
03
33
AM
PDT}

Also in Mung's comment #7:
"...you’ve already assured us that DNA is just simple strings."
Really? And exactly where have I done this?Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
03:29 AM
3
03
29
AM
PDT}

Mung in comment #7: Which of the two strings is more "complex"? You have clearly indicated that neither string is "complex". Or, to state it another way, you have indicated that the two strings have the same degree of "complexity", i.e. none. Now that I have revealed the method by which the two strings were generated, would you like to change your answer?Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
03:27 AM
3
03
27
AM
PDT}

In comment #3 Paul Glem correctly identified the second string: it is indeed simply the first 38 numbers one obtains by counting up from zero in base two (i.e. binary code). This is easier to see if you separate the digits: 0 1 10 11 100 101 110 111 1000 1001 1010 1011 1100 In base ten this would be: 0 = 0 1 = 1 10 = 2 11 = 3 100 = 4 101 = 5 110 = 6 111 = 7 1000 = 8 1001 = 9 1010 = 10 1011 = 11 1100 = 12 If I were to reproduce the second string using the same method, the probability that I would get the exact same string would be 1 (i.e. 100%). I generated the first string by flipping a coin 38 times. If I were to do this again, the probability that I would get the exact same string again would be 1/2exp38 = 1/274,877,906,944 (a very, very, very small number). So, which of the two strings has more "complex specified information"? Or, to ask the same question another way, which of the two strings would require more information to exactly specify its recreation, and why?Allen_MacNeill_{February 20, 2010
February
02
Feb
20
20
2010
03:23 AM
3
03
23
AM
PDT}

And what does this tell us about our ability to detect design in very simple strings of bits (such as are found in a DNA molecule)?
ABSOLUTELY NOTHING! Give us a complex string and we can talk. A complex string from DNA would be even better. But then, you've already assured us that DNA is just simple strings.Mung_{February 20, 2010
February
02
Feb
20
20
2010
01:30 AM
1
01
30
AM
PDT}

Allen:
Is there any detectable design or pattern in either one? To be specific (pun intended), which of these strings exhibits “complex specified information” (”CSI”)? Does one string exhibit more CSI than the other? How do you know? And what does this tell us about our ability to detect design in very simple strings of bits (such as are found in a DNA molecule)?
You know Allen, for someone who spends as much time on the "intelligent design" issue as you do you often display a marked lack of familiarity of the source material. I'd love to be a student in one of your classes on ID.
And what does this tell us about our ability to detect design in very simple strings of bits (such as are found in a DNA molecule)?
So you think ID is about the claim that it is possible to detect design in very simple string of bits? How do you determine that some string of bits is "simple" vs. "complex"?
And what does this tell us about our ability to detect design in very simple strings of bits (such as are found in a DNA molecule)?
So you're claiming that what we find in DNA is nothing more than simple strings of bits?
...which of these strings exhibits “complex specified information” (”CSI”)
I'm going to go way out on a limb here.... You've already stated that these strings are "simple," though you haven't told us how you know that. So you want to know how do we know which of these simple strings contains complex specified information? AARRGGGHHH!!!!!! MY HEAD! You're a university professor aren't you Allen? You have a doctorate? Have tenure? And you want to know how to tell how something which is simple is complex? God help us.Mung_{February 20, 2010
February
02
Feb
20
20
2010
01:27 AM
1
01
27
AM
PDT}

In case anyone's bored: agadaaggbbagaaecaegaaacfbafadafcbgaaaggapyueaaa aaaaaaacgaageaeagaaaaeacagaagggaaaaggaapucnsaaa gaagarunbaaaaabgbaegabaaaadagacaacaafcagaaoqgag aaaaisuvcadgcaafaeabeaaaacpuuzuzurgaaaagghumaaa aeaapuvunggaaagdfaggaaefapcjqanlacapspueguodada faaasscstaaaagfyuuuupgcgaoaohclpbacwuocjupaagba faaevmaqudfacaafaaaagfaaadariamhacapzfdsuzyqaag agchukefymggabfuuvuvpdgaagboeanraaawsgaaaacgabe agaquuyuwqebcaaaaddaabagdabsdgomagapsfcfgaaacga aamusgggmwofgaaaagcgaefagahxagtmekgvtegaaaaddga eatugeaagzugaaagbbdbgaagdcuheaksxjawsaaaaggaaaa 1147Apollos_{February 20, 2010
February
02
Feb
20
20
2010
12:31 AM
12
12
31
AM
PDT}

What has been described with some of the most flowered awe-inspiring words ever used in science, that which facilitates the coordination of all lifeforms in the most complex physicality which we still do not even come close to comprehending, that which required years of investigation just to understand how little we knew, may be carelessly charaterized as a "simple string of bits" in the defense of the othordoxy. In any case... Specific questions asked. Specific answers given. Rebuttal?Upright BiPed_{February 19, 2010
February
02
Feb
19
19
2010
11:44 PM
11
11
44
PM
PDT}

Allen (#2), I am unable to say anything about the first string except that it has more 0's than is likely through chance alone. It may or may not contain a message. The second string is clearly not just non-random, but completely specified. It is simply the binary numbers 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, and 12. It contains 32 digits, with 2^32 possibilities, or slightly more than 4 billion. Although it is technically below the Dembski universal probability bound, if I were a betting man, I would take a million to 1 odds that it was designed, and try to make money. The first string may or may not have complex specified information. One could not be sure it did unless one could find the specificity, although it is a fair bet that it is not random. The second set is clearly precisely specified. It would seem that your point is that we are not very good at detecting design. That's probably true. That has no necessary connection with whether the design is there or not. Awhile back I discussed the matter of whether 1 or 2 lanterns were hung in a tower. The reason this was done was specifically so that design would not be detected. On the other hand, if the British had intercepted a message that said, "Paul Revere, the British are coming by sea", they would have been rather foolish not to have made a design inference. Just because we cannot always make a design inference when design is present, does not mean that the design inferences we do make are incorrect. More to the point of this particular thread, different levels of design can be distinguished. That is why if someone deliberately bashes in a victim's head with a hammer, the hammer manufacturer is not charged with either murder or attempted murder.Paul Giem_{February 19, 2010
February
02
Feb
19
19
2010
09:23 PM
9
09
23
PM
PDT}

How about these two strings: 01000110111000000000101000010000011010 and 01101110010111011110001001101010111100 Is there any detectable design or pattern in either one? To be specific (pun intended), which of these strings exhibits "complex specified information" ("CSI")? Does one string exhibit more CSI than the other? How do you know? And what does this tell us about our ability to detect design in very simple strings of bits (such as are found in a DNA molecule)?Allen_MacNeill_{February 19, 2010
February
02
Feb
19
19
2010
06:22 PM
6
06
22
PM
PDT}

Barry, Those are great analogies. In the first sentence, there is actually design of course -design of the letters, the website, the algorithm you chose to get randomness- but the second sentence has a higher level of design which is detectable by csi (and by intuition). Same thing with the house materials and a finished house. The house materials are at Home Depot by design and are themselves designed for a purpose. That design can be detected as well as the design inherent in the finished house.Collin_{February 19, 2010
February
02
Feb
19
19
2010
01:13 PM
1
01
13
PM
PDT}

Prev 1 … 3 4 5

You must be logged in to post a comment.

Leave a Reply