Wednesday, May 21, 2008

Can Random, Non-Directed Processes Create DNA Information?

In my previous post on DNA, I mentioned the following argument:

A) DNA is information.
B) Information cannot arise from a random, non-directed process.
C) Darwinism requires DNA to have arisen from a random, non-directed process.
D) Therefore, Darwinism cannot explain DNA.
In my first post, I demonstrated A) DNA is information. In this post, I will demonstrate B) Information cannot arise from a random, non-directed process.

The first thing to note is an example that Apolonio brought up. He said:

For example, we can conceive of a case where a person knocks over a scrabble box and the letters I Love You comes out with that order.
While this would be a random process creating information, it is not using foundational forces. The specific example requires a person to knock over the Scrabble box. But even if we adjust for that and make it gravity pulling a box off a shelf or something similar, Scrabble tiles are not foundational in nature; they are designed. So the information still requires a non-foundational force (human ingenuity) to create the tiles which are used to create information in the pattern “I love you.”

Even then, the odds that “I Love You” would form are quite rare. Assuming an equal sample of each letter of the alphabet (as well as an infinite supply of them), you have 8 letters, so the odds of pulling these particular letters would be 1/268, or 1 in 208,827,064,576, which is: 4.79 x 10 -12. If you include the space as a character, we have 10 characters and 27 possibilities each draw: 1/2710, or 4.9 x 10-15.

In reality, however, Scrabble boxes do not contain an equal sampling of each letter. Instead you have 12 Es; 9 As & Is; 8 Os; 6 Ns, Rs, Ts; 4 Ds, Ls, Ss, Us; 3 Gs; 2 Bs, Cs, Fs, Hs, Ms, Ps, Vs, Ws, and Ys; 1 J, K, Q, X, Z. Finally, there are 2 blanks. This yields 100 total pieces. If we use the blanks as spaces, the odds for each letter in “I [blank] Love [blank] you” are:

I = 9/100
Blank = 2/99
L = 4/98
O = 8/97
V = 2/96
E = 12/95
Blank = 1/94
Y = 2/93
O = 7/92
U = 4/91

Because we are not dealing with an infinite number of tiles, we have to reduce how many are available after each selection. Thus, we have a 9/100 chance of pulling an I on the first draw from the box. If we do so, there are now only 99 tiles remaining, 2 of which will be blanks. That means we have a 2/99 shot for the blank, etc. Note that when a letter repeats (for instance, the O), we have to decrease the number remaining too. Thus, the first draw of an O is 8/97 but the second is 7/92 (because the first draw picks one of the Os). Finally, we get the combined odds by the following:

9/100 x 2/99 x 4/98 x 8/97 x 2/96 x 12/95 x 1/94 x 2/93 x 7/92 x 4/91, which is:

774,144 / 62,815,650,955,529,472,000

Or 1.23 x 10-14

Which is roughly 1 in 81 trillion. So even though the tiles were created by humans, a random arrangement of them to spell out “I love you” is still extremely rare.

The above does, however, help us understand a bit about DNA. As most are already aware, DNA uses 3-base codons to create amino acids. There are four possible DNA bases (ACGT), and that means that means 43 (64) possible combinations of those letters. However, there are only 20 amino acids. As a result, amino acids are often encoded by multiple numbers of codons. For instance, Leucine (L) can be encoded by CTT, CTC, CTA, CTG, TTA, and TTG. Which means there are 6 possibilities for L. In fact, quickly going through the amino acids (using their single-letter code name) we find:

I = 3
L = 6
V = 4
F = 2
M = 1
C = 2
A = 4
G = 4
P = 4
T = 4
S = 6
Y = 2
W = 1
Q = 2
N = 2
H = 2
E = 2
D = 2
K = 2
R = 6
Stop = 3

As you can see, all 64 possible combinations would be represented in the above. Therefore, we can say that given a random piece of DNA with 3 codons, there is a 3/64 chance that it is I (Isoleucine) and a 2/64 chance that it is N (Asparagine), etc.

Because base pairs are so prevalent, we can treat them as if there is an infinite supply of them. As a result, if we wanted to calculate what the odds would be that six base pairs will code for Isoleucine and then Asparagine, we would simply multiply 3/64 and 2/64 to yield: 6/4096, or about 1 in 683.

Of course, proteins can have hundreds of amino acids chained together in polypeptides. (In fact, by convention, most scientists do not consider a polypeptide chain to be a protein until it has at least 50 amino acids in it, although that is an arbitrary dividing line.) Because of their size, the odds of even a single 50-amino acid polypeptide forming are quite rare. In fact, even if they were simply a chain of L (Leucine), which has a 6/64 chance of forming for each L, the odds of 50 formulating would be 650 / 6450, which is roughly 8 x 1038 / 2 x 1090 which is approximately 4 x 10-52, or 1 chance in 3 x 1051.

Clearly, this method of explaining DNA is insufficient to explain even a basic protein, let alone complex cells and higher organisms.

This brings us to our next point, which is something that Mighty Pile brought up: the definition of information (i.e., something that is non-repeating, non-random, and not based on foundational forces) seems to exclude the ability of random, non-directed processes in the first place. As such, B) seems to be proven by stipulation, which means it relies on a circular argument.

However, when we examine B) carefully we see that it does not rely on circular reasoning when cashed out. To demonstrate how that is possible, I must first point out that the Darwinist must assert the opposite of B). They must assert that information can arise from random, non-directed processes (as evidenced by premise C)). And this is demonstrated by the fact that you are reading this blog post, which is information.

This blog post has an author. The author is not a random, non-directed process. But, if Darwinism is correct, at some point we can link my existence back to a random, non-directed process. Therefore, in a causative sense, the Darwinist would say that a random, non-directed process somehow created a non-random, directed process that was able to create information.

And it is because this option remains open to the Darwinist that B) does not entail circular reasoning. All the Darwinist needs to do is to show that Information can arise from forces that are non-random, non-repetitive (to exclude crystals) and non-foundational if those forces (we will call them meta-forces) are themselves built on random, non-directed forces. In other words, the Darwinist can argue: “Information comes from meta-forces, which are non-foundational; but meta-forces come from foundational forces.” Putting it into this two-step process would avoid the circular reasoning charge, while also giving the Darwinist a possible route to establishing C).

So the question now becomes, can random, non-directed processes create non-random and non-repeating meta-processes that could then create information in the form of DNA? DNA is one of the simplest information processes we can think of (compare it to trying to establish the framework for a spoken language), but even it is vastly complicated. In order for DNA to function, it has to store information that is used to create amino acids that bond together to form proteins that then create the mechanism for storing and reading DNA. In other words, in order for DNA to function biologically, we need to have a loop where DNA is used to create the processes needed to create more DNA. DNA is copied via cellular processes that are created with proteins that are themselves created by DNA. Thus, we have a vicious cycle going on.

But before we get to the loop, is there a simple way to just encode amino acids into DNA? Amino acids, after all, are fairly easy to create in a test tube, as Stanley Miller demonstrated (albeit his experiment does not prove what he thought it proved). Using those same “primitive” conditions, however, it is not possible to create DNA.

DNA also presents a problem because, as you’ve seen above, sometimes as many as six different DNA codons can represent a single amino acid. While moving from a DNA codon to an amino acid is easy, moving from the amino acid to a particular strand of DNA is much harder.

Due to the limitations of DNA, Francis Crick proposed that life began based on RNA instead of DNA. RNA is only single stranded, as opposed to the DNA double helix. RNA can also sometimes function similarly to proteins. DNA, however, is much more stable and less prone to errors (which is why an intelligent being would pick DNA instead of RNA to start life off; and which is why Darwinists claim DNA was “selected for” by Natural Selection).

Which brings up an important point. The “central dogma” (as Crick named it) is DNA to RNA to protein. It doesn’t go in the opposite order. (There are a few exceptions to the strictness of the “central dogma”, most notably RNA viruses (like HIV) which go from a single strand of RNA to DNA before then going through the “central dogma”; but there are no instances that I am aware of where proteins go to RNA then to DNA.) This makes it highly unlikely that amino acids bonded to become proteins and then those proteins created RNA that was then made into DNA and eventually stored in cells.

That means we had to start someway with DNA or RNA and then create proteins from that; but in order to create the proteins, it means we must have the structure in place by which RNA can be converted to a protein. Once again, we’re left with the chicken and the egg problem. And this system cannot have arisen by blind chance, since as you’ve seen even a single protein of 50 of the most common amino acids has astronomically long odds at forming randomly.

Regardless of where we start, we have to have some method of going from a random soup of amino acids to a particular sequence of amino acids being coded in information, be it RNA or DNA. But this will only start to happen if there is a reason for the information of a protein’s make-up to be converted to RNA or DNA.

That DNA is useful for life is not debated. Suppose that the amino acid “soup” manages to create a protein that could be used by a cell later on. It would be useful for the cell to have a way to rapidly create this protein. And the protein is created from amino acids that can be stored in DNA. Obviously, if we have this end in mind, we could design the process by which the DNA code comes about. But this requires teleology, which Darwinism denies. We cannot have the end of a working cell in mind; we have to have completely random processes that somehow create the necessary steps involved.

But suppose that we are left with only the random creation of the system to begin the evolutionary process. According to modern materialistic theory, life first became possible about 3.5 billion years ago. That is, the Earth cooled enough, the atmosphere was in the correct state, water existed, etc. so that life would not be extinguished if it was formed. Amazingly enough, according to these same scientists, the first life on Earth appeared roughly 3.5 billion years ago. In other words, as soon as it was possible for life to exist on Earth, life did exist on Earth. This must mean that the creation of life ought to be an “easy” process, given materialistic claims. If it is easy, then it should not rely on a process that has such poor odds of succeeding. Either life’s occurrence on Earth was a miracle against all odds, or else this cannot be how life began on Earth.


  1. "While this would be a random process creating information, it is not using foundational forces. "

    This contradicts (B). B states that information **cannot** come out of random process. I don't know in what way are you speaking of cannot here. Metaphysical or logical? It seems that my example, as you said, is a counterexample to (B).

    You said it was not foundational forces. Well, (B) states nothing about foundational forces. And two, it seems to beg the issue that foundational forces must be designed. I think what you are arguing is that information cannot come from non-information. Yet, I think we can think of counterexamples to this.

    I think (B), then, needs to be refined as, "It is improbable that..." rather than saying it cannot.

  2. Apolonio,

    You are correct that my wording was imprecise. In reality, I should have said "While this would be a semi-random process creating information" since the intelligence factors involved mitigate against pure random processes. That is, if it was totally random then the tiles would have needed to be formed randomly too; but they were not.

    I think I explained that part in the paragraph, but the sentence you quoted would be misleading and would contradict my B) statement.

  3. Oh, and to answer your question as to what "cannot" means, I am using it the way scientists typically use it (which isn't a logical impossibility, but rather the probability is so low as to be certain it can never happen).

    For instance, if there are 31,536,000 seconds in a year (60 sec x 60 min x 24 hours x 365 days--not adjusting for leap years) and we say the universe is 15 billion years old, that gives us 473,040,000,000,000,000 seconds (4.7 x 10^17)

    To get the easiest 50-amino acid long protein chain yields odds of 1 chance in 3 x 10^51, which is 34 orders of magnitude larger than the number of seconds the universe has existed (according to modern theorists). Even if we had one combination per second for all 15 billion years the universe is supposed to have existed, we would not have randomly produced a 50-amino acid protein.

    But of course things are bleaker for the Darwinist since they do not have 15 billion years. As soon as life could begin, it did begin. Therefore, you're looking for something that has to be at most a few million years in the making, which in geological time is nothing.

  4. One more comment...

    Even *if* the tiles were designed, the *process* of me knocking it over was still random. If we are going to argue that "foundational forces" are intelligent and therefore the process are semi-random, it seems that there can be no such thing as randomness, which is counterintuitive.

  5. Apolonio said:
    Even *if* the tiles were designed, the *process* of me knocking it over was still random. If we are going to argue that "foundational forces" are intelligent and therefore the process are semi-random, it seems that there can be no such thing as randomness, which is counterintuitive.

    I think you understood me backwards. ;-) It's not that the foundational forces are intelligent (they cannot be) but rather, because you are an intelligent being, then no action that you do is truly random. Of course, it can still be impossible to predict an outcome because the variables are beyond our scope. For instance, if you flip a coin we can treat that result as a random event, but in reality the flip of the coin is determined by the amount of force you apply, the angle the coin is held, how far above the ground it is, wind resistence, and other factors. It is not actually random, but because we cannot predict the outcome (and because we do not have control of many of the variables) we consider the results random.

    One other thing: it is possible for intelligent creatures to create some seemingly random things (i.e., computers can be "seeded" to randomly generate lists of numbers--even then, because they are seeded they are not truly random but only semi-random), but the opposite is so much more difficult as to be physically impossible. Indeed, most of the time when random forces appear to have created something intelligent, it is more an indication of how the human mind can "fill in the blanks" and see what's not really there than it is that the random process actually does such a thing (think Mary appearing in a grilled cheese sandwich).