THE "FATAL FLAW" WITH RICHARD CARRIER'S ABIOGENESIS ARGUMENT

 

The "Fatal Flaw" with Richard Carrier's 

"1 in 10^41" Argument for Abiogenesis

(See, "Q's Response to Richard Carrier: About the Debate," for context)

(See also, "Q's Open Challenge to Richard Carrier")

Q's Verdict (Summary) 

Richard Carrier’s central, foundational “1 in 10^41” argument for abiogenesis that he has repeatedly asserted for twenty years is absolutely, indisputably, and unequivocally false, because it is based on a faulty understanding of how a "Lee peptide" self-replicates. He believes, for example, that a "Lee peptide" once formed in the "right environment" will then "start spontaneously creating copies of [itself]...as a consequence of known physical laws" by "attach[ing] amino acids" together. But that is not how a "Lee peptide" self-replicates. It does not have the ability to chain amino acids together at all in any order, much less the correct order (even in the "right environment"). It self-replicates in a very specific way by taking what is effectively a second identical "Lee peptide" copy of itself that is 'broken' in half, and favorably positions the two halves so they can connect themselves together (provided they have been chemically altered in the right way to do so). Richard would know this if he had carefully read his primary source citation. Thus, in order for self-replication to occur, we not only need the spontaneous origin of a "Lee peptide" (1 in 10^41), but also need what is effectively a second identical "Lee peptide" copy ('broken' in half) to spontaneously form in the same place (another 1 in 10^41 = 1 in 10^82), so our first "Lee peptide" can help the two halves connect together. We then need another set of "Lee peptide" halves to spontaneously form in the same place (1 in 10^123), and then another (1 in 10^164), and another (1 in 10^205)—a continuous supply—in order to have a self-replication reaction that is self-sustaining. The instant we run out of "Lee peptide" halves, the reaction stops. This is but one of a multitude of problems with Richard's arguments. (See also, "Q's Open Challenge to Richard Carrier").

Richard Carrier's Argument for Abiogenesis 

Richard Carrier claims that 1 in 10^41 is a minimal probability of abiogenesis; that this is “so far within the realm of cosmic possibility that it is already certain to have happened many times [even within the visible universe]”; and that therefore, this effectively ‘proves’ a naturalistic origin of life with 100% certainty. Richard claims the true probability is actually better than this when we factor in the sum total additive probabilities of “countless other initial self-replicators…any one of [which could arise] by chance,” but he gives 1 in 10^41 as a minimum starting point. He says this is the probability of "the random assembly of the simplest known self-replicating peptide molecule"—a 32 amino acid long polypeptide that Richard dubs the "Lee peptide" after the primary source (Lee et al. (1996). A self-replicating peptide.). He credits McFadden in his 2002 book Quantum Evolution for "the reasonable calculation" based on the correct sequencing of twenty canonical amino acids in each of thirty-two positions (i.e., 20^32 = 10^41) (Note: Richard says random assembly is actually "more probable" than this, but see my response). Richard further claims that he accounts for “all other components” necessary for biogenesis, but says the chance assembly of an initial self-replicating molecule is the only "hard part," and thus none of these other factors “have any significant effect on that [1 in 10^41] probability”. 

In Richard's view, the origin of an initial self-replicating molecule is the key, defining moment—the pivotal event—in abiogenesis that is needed to "spark" life, and "it's getting that first spark that's the hard part". It is the critical bottleneck in chemical evolution. In his mind, there are "no extraordinary improbabilities" beyond this single hurdle—either before, or after—and once formed (in a "conducive environment"), an initial self-replicator will "start spontaneously creating copies of [itself]...as a consequence of known physical laws" by "attach[ing] amino acids to a base or backbone of some other chemical to maintain their order and structure". It is at this point, according to Richard, that we have a "naturally self-sustaining chemical system" where "evolution by natural selection is then operating," and can experience rapid, exponential growth; where, for example: "[a] single replication per day (i.e., one copy making just one copy in the course of one day) results in a million copies in just 27 days; another 27 days, and you are at 10^12 copies; even before you get to a full year you're already at 10^81...That's how powerful evolution by natural selection is."

In Richard's view then, it's just "getting that first spark that's the hard part," and that "hard part" requires at most only beating the 1 in 10^41 odds of spontaneously forming a 32 amino acid long "Lee peptide" by chance. Based on the Dembski standard of 1 in 10^150 that Richard cites:

"[E]vents as unlikely as 1 in 10^41 will have occurred by random accident in this universe over 10^109 times...Even if a tiny collection of peptides came together in a suitable environ only once ever, across all trillions of galaxies after billions of years, and chained only once, the probability that something this unlikely has happened somewhere in the universe by now is essentially 100%. Something of the like is guaranteed to have happened." ("Biogenesis and the Laws of Evidence")

This "1 in 10^41" probability is, thus, the core, foundational bedrock of Richard's argument for abiogenesis; an argument that he has repeatedly asserted for twenty years, going back to his 2004 B&P article, and in numerous blog entries (See, for example, here, here, here, here, here, and here; from which the above quotes are derived, and wherein he references his "1 in 10^41" argument at least twenty-five times).

Richard does not understand how a "Lee peptide" self-replicates

Richard is, of course, wrong, first and foremost, because he does not understand how a "Lee peptide" self-replicates (He is also confused about what a "Lee peptide" is; see, "Richard's First Mistake: a 'Lee peptide' is not a PNA")  It does not have the ability to chain amino acids together at all in any order, much less the correct order (even in the "right environment"). Thus, a "Lee peptide" in the "right environment"—or even an idealized environment with an unlimited supply of activated amino acids in some "warm little pond"—will 'sit' and do nothing. Richard would know this if he had carefully read and understood his primary source citation.

A "Lee peptide" does not have the ability to chain amino acids (even in the "right environment")

A "Lee peptide" self-replicates in a very specific way by taking what is effectively a second identical "Lee peptide" copy of itself that is 'broken' in half, and favorably positions the two halves so they can connect themselves together (provided they have been chemically altered in the right way to do so). More precisely, the "Lee peptide" adopts a coiled, corkscrew-like shape known as an alpha-helix that enables it to intertwine with a second alpha helix like the strands of a rope, and form a structure called a coiled coil.  

A coiled coil formed by two alpha helices

This coiled coil structure is the mechanism by which a "Lee peptide" self-replicates. The "Lee peptide" forms a coiled coil structure with the two halves of a second identical "Lee peptide." Specifically, a seventeen amino acid long 'half' (17aa) and a fifteen amino acid long 'half' (15aa) that have been chemically altered in a very specific way to promote their joining. 

17aa 'half''                                    15aa 'half'

32aa "Lee peptide"

Amino acid sequence of a "Lee peptide" and its corresponding 17aa & 15aa 'halves'
(See, "Template", "Electrophilic fragment," and "Nucleophilic fragment", Figure 2, Lee et al. 1996)

A "Lee peptide" forms a coiled coil structure with two 'halves' of a second "Lee peptide"
(See, Figure 5a, Lee et al. 1996)

A "Lee peptide" does not have the ability to join the halves together. In order to connect, the two halves must first be energetically activated in a very specific way via a complex chemical process that is not prebiotically plausible (but that is another problem entirely), and then must be in close enough proximity and facing the proper way in order to form a chemical bond. A "Lee peptide" accelerates (i.e., catalyzes) this process by forming a coiled coil structure that does precisely that: i.e., it physically orients the two chemically altered 'halves' the correct way and in close enough proximity to each other so they can chemically bond themselves together. Although, 15% of the time the two halves still do not connect together the right way (but that is also another problem entirely). Again, Richard would know all this if he had carefully read and understood his primary source citation. 

(See, Figure 1, Lee et al. 1996)

Like most catalysts, the "Lee peptide" is substrate specific and can only catalyze one type of chemical reaction. The reactants in an enzyme catalyzed reaction are called substrates. For example, in the reaction below, A and B are the substrates that enzyme catalyst C acts upon to make product D: 

A + B —[Enzyme catalyst "C"]—> D

The only difference between this and self-replication is that a self-replicator molecule catalyzes a reaction where the product is an identical copy of itself:

A + B —[Self-replicator catalyst "C"]—> C

The specific substrates in the self-replication reaction catalyzed by a "Lee peptide" are the two (17aa & 15aa) halves of a second identical "Lee peptide":

 

The "Fatal Flaw" with Richard's "1 in 10^41" Argument

This brings us to the "Fatal Flaw" with Richard's "1 in 10^41" argument for abiogenesis. Indeed, it is the "Fatal Blow"—the "Death Stroke." Quite simply, Richard forgets about the substrates: a catalyst (even a self-replicating one) without any substrates has nothing to do! (i.e., no reaction to catalyze). A self-replicator cannot make more copies of itself if it does not have the requisite 'building blocks' (i.e. substrates) from which to do so; and in this case, the substrates are the two (17aa & 15aa) 'halves' with identical amino acid sequences as our "Lee peptide."

The "Fatal Flaw" with Richard's "1 in 10^41" argument for abiogenesis

Thus, even if we beat the 1 in 10^41 odds and spontaneously form a "Lee peptide" self-replicator in the "right environment," nothing will happen, because our "Lee peptide" has nothing to do. It has nothing to do, because it has nothing to catalyze, so it will 'sit' around and do nothing (at least until it decomposes back into its component parts). It cannot self-replicate unless it has substrate “building blocks” (i.e., the 17aa & 15aa 'halves') to help connect together to make more copies of itself.  That is why the spontaneous formation of a "Lee peptide" (even in the "right environment") is not sufficient for abiogenesis. We also need the 17aa & 15aa halves (with identical amino acid sequences as our "Lee peptide") to also spontaneously form in the same place, so our "Lee peptide" can help connect the two halves together to make another identical "Lee peptide" copy of itself. We then need another set of 17aa & 15aa halves to spontaneously form in the same place, and then another, and another—a continuous supply—in order to have a self-replication reaction that is self-sustaining. The instant we run out of 17aa & 15aa halves, the reaction stops. 

Why abiogenesis is far, far, far more improbable than "1 in 10^41"

But what is the probability of all this happening? If we employ McFadden's same method of calculation that Richard relies on to get his "1 in 10^41" odds, then we get the following:

17aa 'half": 20^17 = 10^22 

15aa 'half': 20^15 = 10^19

The probability that our 17aa & 15aa 'halves' will both spontaneously form somewhere in the universe is the product of their probabilities:

(10^22) x (10^19) = 10^41

Notice that this is the same as McFadden's 1 in 10^41 probability for the spontaneous origin of a single 32aa “Lee peptide.” Thus, the probability of spontaneously forming a 32aa "Lee peptide" and our 17aa & 15aa 'halves' (so we can have just one single round of self-replication) is equivalent to spontaneously forming two “Lee peptides”:

(10^41)^2 = 10^82

Now based on the Dembski standard of 1 in 10^150 that Richard cites, 1 in 10^82 is still within the realm of possibility. Thus, we can say 'on paper' that somewhere in our 'toy model' universe there is a 32aa “Lee peptide” and 17aa & 15aa "Lee peptide" 'halves'. But, of course, this is just the probability of all three spontaneously forming somewhere in the universe. It is even more improbable that all three will form at the same time and place.

But let’s say it happens anyway, and our “Lee peptide” connects the 17aa & 15aa 'halves' together to make an identical 32aa “Lee peptide” copy of itself (which it doesn't actually do, and which 15% of the time don't connect in the right way). What happens next? Answer: Nothing. We now have two “Lee peptide” self-replicators 'sitting' around with nothing to do, because we are out of substrate “building blocks” again.

So now what’s the probability of forming another set of 17aa & 15aa 'halves'?

Answer: (10^41)^3 = 1 chance in 10^123.  And that is the limit of what we can expect, as (10^41)^4 = 10^164, exceeds Dembski's 10^150.

Of course, once again, that’s just the odds of all five molecules spontaneously forming somewhere in the universe (i.e., one “Lee peptide”, and two sets of 17aa & 15aa “halves”). The chance that all five would spontaneously form in the same time and place is far, far, far more improbable.

But let’s say it happens anyway, and that one or the other of our two “Lee peptides” catalyzes a second round of self-replication. What happens next? Answer: Nothing. We now have three “Lee peptide” self-replicators 'sitting' around doing nothing, because once again we are out of substrate “building blocks.”

However, if we are going to have a self-replication reaction that is "naturally self-sustaining" and grows exponentially, then the problem becomes exponentially worse. A single round of self-replication requires the equivalent of two "Lee peptides" (1 in 10^82). A second round requires the equivalent of two more "Lee peptides" (1 in 10^164) in order to double our number of self-replicators from two to four. To double that number again to eight with a third round of self-replication, requires the equivalent of four more "Lee peptides" (1 in 10^328). A mere ten rounds of self-replication to give us 1,024 "Lee peptides" mushrooms to the absurd improbability of 1 chance in 10^41,984!

Beating the "1 in 10^41" odds and spontaneously forming a "Lee peptide" is not enough. A continuous supply of 17aa & 15aa 'halves' must also spontaneously form in order to have sustained self-replication of a "Lee peptide."  

Conclusion

To borrow a line from Richard, "So we're done, really. I could drop the mic here." For even if we get that "first spark"—i.e., the "hard part"—by beating the "1 in 10^41" odds and spontaneously form a "Lee peptide" in the "right environment" somewhere in the universe (or even do it multiple times), it turns out there's an even "harder part": spontaneously forming a continuous supply of substrate 'building blocks' (i.e., 17aa & 15aa 'halves') by random assembly, which is impossible by any standard.

The truth is we've only just started going down the "rabbit hole." The problems only get worse from here the more realistic we make things (See, for example, "The 'Concentration Threshold' Problem," and "Why the 'Lee peptide' Cannot Sustain Exponential Growth Even with a Continuous Supply of Substrates"). Richard's other self-replicator examples are plagued by similar difficulties, which deserve to be taken in turn. But it is enough and will suffice for now to simply reassert the truth with which we began:

Richard Carrier’s central, foundational “1 in 10^41” argument for abiogenesis that he has repeatedly asserted for twenty years is absolutely, indisputably, and unequivocally false.