In a reply to Gordy Slack, Nick Matzke wrote something great at the Panda's Thumb on the same topic and I want to paste in the relevant portion below:
An excellent summary, if I may say so. He left out a few things that I've mentioned before, like the discovery of the role of borate in natural ribose synthesis by a colleague of mine at UF, and upstairs neighbors, Alonso Ricardo and Fabianne Frye. Go here for some great peer-reviewed publications on the origin of the genetic code and the evolution of the chemical processes underlying metabolism.Origin of Life (OOL)
Slack lists a “few worthy points” creationists make. Here is the first:
First, I have to agree with the ID crowd that there are some very big (and frankly exciting) questions that should keep evolutionists humble. While there is important work going on in the area of biogenesis, for instance, I think it’s fair to say that science is still in the dark about this fundamental question.
Minor point first: Shallit points out that “biogenesis” means production of life from life, whereas Slack is talking about the origin of life (OOL). Oops.
Major points: PZ says that sure, big exciting unanswered questions like the origin of life exist in science, but scientists said this first, and furthermore consider them research opportunities, not flaws. Shallit separates OOL from evolution, specifying that evolution is what occurs after you have life; Shallit does this in the face of Slack asserting that this response is disingenuous. Shallit also argues a little over whether or not we’ve made only “little progress” in understanding the OOL, but says even if we’ve only made a little progress, it’s better than ID.
This mini-debate points out what I think, and have often said in conversations, is a major flaw in how we respond to creationists. All too often, when the OOL comes up in popular discussions (reporters, online debates, etc.), the anti-creationist will reply with some variation of “sure, it’s a tough unsolved problem, but we’re working on it”, or the wizened statement “actually, the OOL is outside of the domain of evolutionary biology”, or finally, “we’re pretty much in the dark about the OOL, but at least what we have is better than the creationists giving up and saying a miracle occurred.”
My take: It is high time all of these statements be discarded or highly modified. They are basically lazy, all-too-easy responses relying on hair-splitting technicalities or nearly philosophical assertions of the “even if the creationists were empirically correct on this point, which they aren’t but I’m too busy to back it up right now, it wouldn’t matter” variety. And the worst part is that these sorts of statements mis-describe the actual state of the science among the people who work in the field. It is simply not true that we, the scientific community, know almost nothing about the OOL (what an individual who spent a career working on fossils or fruit flies or speciation might know personally is a different question).
Here is a short list of things we have discovered or confirmed in the last 50 years or so pertaining to the origin of life. In my opinion all of these points have reached high enough confidence that they are unlikely to change much with future discoveries, and our confidence in them does not depend in uncertainties in the remaining unanswered questions.
OOL Discovery #1. All known life can be traced back to a single common ancestor which, compared to what most people think of as present-day life (i.e. plants and animals), was relatively simple – microscopic, single-celled, perhaps as complex as an average bacterium or perhaps somewhat less so.
Because a lot of creationists, and sometimes others, are a bit thick in the head on correctly understanding this point, let me bash away at some common misconceptions. The phrase “single common ancestor” does not, and never has for people who were paying close attention, referred to a literal single individual organism. Think about a phylogenetic tree with humans and chimps on the branches. When you trace the tree back to the “common ancestor” of chimps and humans, does that node represent a literal single individual? No, of course not! Everyone (well, everyone paying attention) realizes that that ancestral node represents a species or population sharing genes in a gene pool. Ditto for all of the other ancestral nodes in a phylogenetic tree, including the Last Common Ancestor of known life.
With this understood, the debate initiated by Ford Doolittle and others over the precise nature of the Last Common Ancestor – they argue that it was a population of unicells that were rampantly trading genes – can be put in the correct context. It’s basically a debate about how wide or narrow the bottleneck the Last Common Ancestor represents, and whether (for example) modern life might contain some genes derived by lateral transfer from pre-LCA lineages that are now extinct. These debates are fascinating and highly technical, but they don’t undermine at all Point #1. Somewhat ironically and counterintuitively, those who say that there was rampant lateral transfer – this is supposed to be the “radical” position that “uproots the Tree of Life” when its proponents get their blood up – are actually pushing the LCA to something more and more like a traditional gene pool, i.e. species, i.e. what every other node in a phylogenetic tree represents.
Any way you slice it, all known life (with minor derived exceptions, and excepting viruses) shares a suite of protein and RNA genes, a DNA-RNA-protein system and a mostly standard genetic code (again with minor derived exceptions), etc. Even if various other bits of modern life came from other ancestral lineages (unlikely for most features in my opinion but there may be some exceptions), this shared system indicates that all known life, i.e. all the stuff that’s not extinct, descends from a pretty good bottleneck where these features were fixed in the “population.” And this reconstructed ancestor is maybe as complex as a typical bacterium and probably less so. It could be that in the last 50 years science discovered that known life had for-real multiple origins, or that at the root of the tree was a complex multicellular organism with 30,000 genes and huge, elaborately regulated, genome, but instead we get a unicell with a relatively small & simple genome. Various caveats, important to scientists but irrelevant to beginner-level education and dealing with creationists (e.g., somewhat more genes may have been passed through the bottleneck in some but not all organisms if the LCA was more of a gene-trading community) should not be allowed to distract from the Main Point: science has confirmed the hypothesis, going back at least to Darwin, that the ancestor of modern life was much less complex than life today.
OOL Discovery #2. The Last Common Ancestor itself was the product of evolution from an even simpler ancestor. The simplest piece of evidence for this is that a number of the genes found in the Last Common Ancestor are homologous, thus derived from a single common ancestor by duplication and modification. An example is the F1Fo-ATPase of bacteria, which interconverts proton gradients (or sometimes sodium ion gradients, which are chemically very similar) and ATP, the main energy currency of cells. It has relatives in all branches of life: the V1Vo-ATPases in eukaryotes (and some prokaryotes), and the A1Ao-ATPases of archaea, and phylogenetic analysis indicates that this membrane-embedded system was found in the LCA (this also confirms that the LCA had membranes, which is useful although already very likely on multiple grounds).
Anyway, the bit of the V/F/A-ATPases that deals with ATP is a heterohexamer, i.e. complex of six proteins (that’s the “hexamer” bit) of two different types (thus “hetero”), cleverly named alpha and beta. The alphas and betas alternate in the six-protein ring, and the betas interact with ATP. The key point here is that the alpha and beta subunits share statistically strong sequence similarity. The simple explanation is that the heterohexamer was descended from a homohexamer made up of six identical proteins forming a ring. Thus we know – as strongly as we know that two people are related by ancestry based on DNA sequence similarity – that long before the last common ancestor of life there was a cellular organism that had something like the F1Fo-ATPase, but a simpler version with a homohexameric ATPase complex instead of a heterohexamer.
This may seem like a trivial point by itself but it is just an example; there are dozens, perhaps hundreds, of others. The evolution of ATPases can be traced much further back: the next closest relative is a homohexamer found in, of all things, the core of the bacterial flagellum and the nonflagellar type 3 secretion systems. Thus the V/F/A-ATPases and the flagellar/nonflagellar type 3 secretion systems can be traced back to an ancestral membrane-associated complex with multiple shared proteins (because the V/F/A-ATPases and type 3 secretion systems shared not just the ATPase protein but also an associated external stalk protein, FliH/Fo-b, which by the way is something I pretty much predicted in 2003 in the Big Flagellum Essay and which Mark Pallen and colleagues nailed down for real in the peer-reviewed literature in 2006).
Even more distant relatives are known: the homohexameric rho (involved in bacterial RNA processing) and homohexameric RecA (DNA processing). And there are even more distant hexameric sister groups; the whole related set of proteins is known as the AAA ATPases if you want to look them up. And if memory serves there are yet more distant non-hexameric relatives.
In other words, the Last Common Ancestor had a suite of ATPase proteins which had already evolved from a single protein ancestor by duplication and divergence events which are still strongly statistically detectable in the present day. And there are a number of other genes for which the same can be said, and undoubtedly many others which occurred but are not statistically detectable billions of years later due to the decay of the signal.
So far we’ve established that anyone, creationist, evolutionist, or whomever, who says that the scientific understanding of the origin-of-life is chemicals –> mystery –> modern-complexity life doesn’t know the first thing about what they’re yapping about. At the very least we’ve got chemicals –> mystery –> quite simple precursor to the LCA –> LCA –> modern life. But there is yet more that we know
OOL Discovery #3. DNA/RNA/protein-based life was preceded by something even simpler, an RNA world or at least an RNA-heavy world.
The RNA world has gotten better press attention than OOL Discoveries #1 and #2 so I will spend less time on it. Read the wikipedia page for an introduction and particularly the EvoWiki page for some of the main supporting evidence.
There are two points worth making about the RNA world that should be made every time this issue is discussed in popular or remedial creationism-related discussions. First: there was a time, not too long ago, when the fact that DNA coded for proteins, and proteins were necessary for making DNA, seemed like the ultimate intrinsically unsolvable problem in the study of the origin of life. It’s the ultimate chicken-and-egg problem, or, if you like, the ultimate “irreducible complexity” problem. And yet, scientists worked on it for a few decades and discovered a workable, surprisingly simple solution.
Second, surprisingly enough (well, surprising to creationists or the naive), this solution, the RNA world, hasn’t just sat around as a purely theoretical just-so story. A highly productive research program has been built on the RNA World concept. Areas that have experienced substantial success in the last decade or two include: the discovery of increasingly diverse catalytic capabilities of RNA; the evolution new capabilities in replicating, evolving RNAs; the evolution of the genetic code which translates DNA to RNA to protein; and the prebiotic origin of RNA components.
Each of these areas has developed into a subfield which has experienced major research discoveries in recent years. For example, on the origin of the genetic code, this paper assembles dozens of indicators on the order in which amino acids were added, step-by-step, to the genetic code and shows that the evidence strongly supports a fairly specific scenario (which shares many similarities with early, more speculative scenarios built on the basis of just a few lines of evidence).
Ergo, we don’t just know that the Last Common Ancestor of Life was simple, and that it’s ancestor was simpler, and that it’s ancestor was an even simpler RNA-dominated critter; we even have a decent idea about the order of the steps by which the genetic code itself evolved.
OOL Discovery #4. The increasingly simple ancestors of modern life weren’t made out of just anything, they were made out of chemicals that just happen to be generated by plausible abiotic mechanisms found in early solar systems. This area is also better known, but many, both creationists and scientists and journalists who haven’t thought about it enough, tend to think of prebiotic chemistry as the beginning and ending of origin-of-life studies, and for some extremely foolish reason which I can’t fashion, probably simple carelessness, tend to think that until chemists pop life out of a test tube then we “know nothing” about the origin of life.
Here’s a short list of discoveries about prebiotic chemistry, all of which increase our confidence in the idea that the origin of life was a gradual process, from abiotic chemicals to simple replicators to the simple ancestors of modern life which were discovered above. I’ll include some subtleties that I’ve seen lead people astray on occasion.
* Water is one of the most common compounds in the universe, and was/is common in the solar system (subtlety: most of it is frozen, but remember that on any planet with hot stuff inside and cold frozen stuff outside will have a just-right region in-between where water will be liquid)
* Earthlike planets are likely reasonably common (subtlety: we haven’t discovered them directly yet, but this is isn’t because they aren’t there, it’s because our instruments are at present only sensitive enough to detect big, close-in planets around other stars. Nevertheless, the distribution of the stuff we can detect strongly indicates that there are plenty of earthlike planets in earthlike orbits which will be discoverable in the near future. That’s a prediction, scientists will test it, that’s science for you. Remember that back in the 1990s, ID proponent William Dembski was skeptical of the whole idea of extrasolar planets. Whoops!)
* Amino acids are easy to generate by a variety of processes, and this is not only supported by experiment, but by observation of amino acids in meteorites and other extra-terrestrial material. (Subtlety: There is a body of serious scientific thought which suggests that the Earth’s early atmosphere was more neutral and less reducing than was thought a few decades ago, but (1) this isn’t for sure, the redox chemistry of the Earth’s rocks and atmosphere is a complex business (and I wonder if the impact which produced the moon, removing much of the mantle but leaving the Earth enriched in heavy iron might have made the Earth’s atmosphere more reducing, at least early on – comments?); (2) even in a neutral atmosphere/ocean system there will be locally reducing conditions – heck, there are local reducing conditions here on earth right now even with our heavily oxidized crust and atmosphere; (3) as it turns out, even neutral atmospheres can produce amino acids in respectable yields anyway; and (4) this whole sub-debate is somewhat moot since we have direct evidence of amino acids forming in the solar system e.g. in meteorites.)
* RNA precursors are somewhat tougher, but there has been progress in that area also, and anyway there is no requirement that the first replicator must have been RNA; various other simpler “worlds” have been suggested and are being explored (PNA, peptide nucleic acids; other NAs of various sorts; and lipid worlds, which have the distinct charm of instant replication ability and statistical inheritance, with daughter bubbes containing a subsample of the chemicals making up the mother bubbles, and growth occuring by incorporation of lipids from the environment and other bubbles; so maybe the first “replicators” were even simpler than some have thought).
* The main energy source of present life is ATP and other energized phosphate molecules. So, what was the prebiotic source of those? It turns out that inorganic polyphosphates (chains like phosphate-phosphate-phosphate-phosphate) have energetic bonds very similar to those of ATP (which is adenine-phosphate-phosphate-phosphate), and yet can be formed by the simple heating of certain rocks.
* Less well-known is the fact that prebiotic origins of many cofactors and other universal small biotic compounds have also been reconstructed
What is actually being worked on. The above should convince you that the idea that we know nothing or very little about the OOL is just uninformed foolishness. The field has made major progress. There are some famous puzzles remaining, but they do not add up to “we know nothing about the origin of life.” Furthermore, some of the puzzles that creationists, and sometimes others, consider to be major hangups, are not necessarily so. For example:
* The origin of chirality (the left-handedness of amino acids). This is a major puzzle if you make the extremely foolish and unthinking assume (like creationists do, but sometimes others) that the first use of amino acids in early life was supposed to be in long amino acid chains made up of 100+ amino acids randomly assembled from an even mixture of 20+ different amino acids with an even mixture of right- and left-handed amino acids. But over here in the real world, where the origin of the genetic code has been reconstructed in some detail, we know the following: the first primitive genetic code used just one or a few amino acids, and one of the first was glycine, which is the simplest amino acid, the most common amino acid produced in prebiotic experiments, and which is achiral (no left-hand/right-hand difference) to boot. If, as has been proposed, the first use of amino acids was as something relatively prosaic, i.e. a short chain of hydrophobic residues to insert into an early membrane, then (a) the odds of getting 10 or so amino acids at once that were either left-handed or glycine were not small at all, and (b) it wouldn’t have mattered much if the occasional right-handed amino acid was incorporated, because the crude chemical property of hydrophobicity is all that is really important, and (c) therefore the origin of a preferred chirality could have been more or less random. There is some very interesting work indicating that nature has various processes which might increase the proportion of left-handed amino acids, but it’s not at all clear that these will be necessary to explain chirality.
* The origin of the first replicator. This really is the big cahuna of the OOL discussion, and where the big and contentious debates are still occurring within science, but again I find that many discussants operate with very crude and naive assumptions about what early replicators “should” have been like and what prebiotic experiments “should” be able to produce to “solve” this problem. It’s a mini-version of the “produce a modern cell in a test tube for me or you haven’t solved the OOL” silliness, i.e., “produce a self replicating RNA World, with duplicating ‘informational genetic sequences’ in the test tube, and until you do you can’t say we know anything about the origin of replicators.”
Again, over here in reality-land the distinctions between replicators and nonreplicators are not so clear. I have already mentioned “lipid-world” ideas and the concept of “statistical inheritance”, where overall chemical properties are transmitted or accumulated, without the need for exact inheritance of a sequence. Similar concepts have been applied by OOL workers to amino acid and nucleic acid “sequences”, where before exact inheritance of sequence is acheived, there might have been a stage where inexact incorporation of a range of chemically similar bases was occurring.
Another subtlety is the difference between “self-replication” and processes where prebiotic compounds go through a series of chemical processes, and differences in chemical kinetics increase the frequency of compounds that have more rapid kinetics; if these compounds are auto-catalytic, they can begin a feedback system where chemicals with higher kinetics take over in a proto-selection system. Strangely, although everyone who takes college chemistry learns that the product of chemical reactions is a combination of thermodynamics and kinetics, many discussions of the OOL from scientists, and all of the derivative critiques by creationists, have focused on thermodynamics. This is particularly odd since self-replication is the ultimate example of kinetics overwhelming thermodynamics.
What’s the point of this sub-discussion? Well, if it is the case that the origin of the first “replicator” was, like everything else we’ve discovered in the study of OOL, a slow, gradual (meaning step-by-step), cumulative process, then it is pretty foolish to have in our heads the idea that OOL experiments should produce full-on replicators in one go to be successful experiments. This is basically a strawman expectation that expresses conceptual confusion about what an evolutionary origin of life “should” look like.
(As an aside, I think biology education would be a lot better off if the above points were consistently made in science curricula and textbooks at the high school and college level. Teaching OOL as a story from simple to complex, rather than a detective story from complex to simple, is probably a mistake if the goal is to get students to understand why scientists think the way they do about these issues.)
The Main Point
Now that we’ve briefly reviewed the OOL field and discussed the major discoveries and some of the common misconceptions, let’s return to the statements I quoted at the beginning. Is it really true that “science is still in the dark” on the OOL, as Slack said? Not a chance. If we lived in a world where it actually looked like the first living things were as complex or more complex than life today, or where the last common ancestor contained absolutely no evidence of an evolutionary history, or where big obvious puzzles like the interdependency of DNA/RNA/protein had no hint of solution, or where the building blocks of life were completely unrelated to those produced in prebiotic experiments – all of these things would be true, say, on a robotic planet without microscopic life, where robots were replicated by macroscopic assembly performed by other robots, and powered by hooking up to a grid of fusion-fueled power plants – then we could say “science is still in the dark” on the origin of this robotic biosphere. But instead, we have numerous lines of evidence all pointing towards the notion that current life descends from a relatively simple ancestor, and that ancestor descends from a series of even simpler ancestors. Why should any of this evidence exist, if life was poofed into existence all in one step, which is what the creationists/IDers think happened even when they won’t admit it, because they are not brave enough to defend what they actually think? Additionally, why should the remaining puzzles, particularly about the origin of the first replicator, cause any unusual amount of discomfort for scientists? Whether or not that puzzle is solved, the gap between prebiotic experiments and the first replicators (or better yet, pseudoreplicators with statistical inheritance) is a drastically reduced vestige of a gap compared to what the gap looked like in, say, 1950. When you think about it, the creationists’ attempt to insert miraculous divine intervention into this tight little gap which is left is actually pitiful, and a pretty sad commentary on the state that creationism/ID has been reduced to. The verse “And God said, let the NA precursors link together into a short noncoding kinetically favored chain and pseudoreplicate approximately statistically after their kind” just doesn’t have the same ring to it.
Similarly, if my characterization of the state of the science is accurate, then it is highly irresponsible for scientists to address creationist arguments about the origin of life with statements like, “even if the creationists were empirically correct on this point, which they aren’t but I’m too busy to back it up right now, it wouldn’t matter” or “actually, the OOL is outside of the domain of evolutionary biology.” The first statement surrenders without argument a favorite bogus creationist talking point, and so confirms and passes on their misinformation, even if the evolution “wins” the argument in his own head on some broader philosophical point. Instead of putting the creationist back on his heels with a wave of contradictory evidence, that sort of response, even if the philosophical point is valid, leaves the creationist and any of his sympathetic readers irate that the empirical point is not being addressed, and that the creationist/ID position is being excluded by the rules of the game. The fact that this sort of response is a lot easier and faster to put together does not make it the best one.
The second statement, splitting the OOL from evolutionary theory, is only technically correct in a sort of legalistic, hairsplitting way. Sure, it’s true that technically, “evolution” only happens once you have life, or at least replicators, but getting from replicators to the last common ancestor is most of what most people think about when they’re thinking about the origin of life, i.e., “where did the evolutionary ancestor of all life today come from?” and all of that is evolution all the way. Furthermore, even the origin of the first classical “replicator” was itself very likely an evolutionary process, in that it occurred in stepwise fashion and not all-at-once, and that the first replicator was likely preceded by various sorts of pseudoreplication, statistical inheritance and kinetic biases. If you remove evolution from your thinking about the origin of the first replicator then it is very likely you will never understand how it happened, or what the current research on the question is about. Finally, even apart from these detailed considerations, “evolution” reasonably has a broader meaning – the evolution of the universe, the solar system, the planet, and the planet’s geochemistry, and the origin of life and the origin of the first replicator must be understood as part of that larger evolutionary history.
One other telling point is that the statement “but the OOL is outside of evolutionary theory” response also has the problem of simply dodging the hard work of describing the discoveries and work of modern science, a problem I have already described. In conclusion, if it were up to me, I would completely scrap this statement from the rhetorical toolkit of evolution defenders.
The OOL topic turned into an essay on its own, but we still have another few of Slack’s points to address.
