Signature in the Cell and Thought Experiment About Design
July 20, 2010 — 13:32

Author: Andrew Moon  Category: Books of Interest Existence of God  Comments: 21

SO, I’ve been reading Stephen Meyer’s Signature in the Cell, I’m about three quarters through, and I have to say that it’s been an enjoyable read so far. Just as fun has been my reading group for this book with two biology majors, a biochemistry major, and a biological engineering major; they help explain the biology concepts to me.
The book’s undergone a lot of controversy; I was motivated to read it primarily by Thomas Nagel’s (in)famous recommendation. There’s a lot to criticize about the book, and people have pointed to possible errors in biology that I’m currently not in the position to assess. There are also some little errors. For example, Meyer writes, “It follows that mind–conscious, rational intelligent agency–what philosophers call “agent causation,” now stands as the only cause known to be capable of generating large amounts of specified information starting from a nonliving state” (341). Of course, this is not what philosophers call “agent causation”, so it’s an error, but it’s a “little error” because nothing of significance falls on it. (In my opinion, most of the critical blog posts on the net that I’ve read so far focus on little errors and not at all on Meyer’s overall argument for ID.) Also, the book is longer than it needs to be, although I often enjoy his excursions and illustrations.

But like the Buddha said, don’t just look at the finger, but at the awe-inspiring beauty that the finger is pointing to. And awe-inspiring the cell is with its magnificent complexity; DNA’s ability to replicate itself — through the whole exquisite translation/transcription process, which requires the existence of proteins, which in turn require the existence of DNA — is remarkable and defies summary.
And from what I can make of things, Meyer shows (or summarizes how others have shown) that none of the current theories that appeal only to non-intelligent causes can account for the existence of ‘specified information’ in such systems. We know that intelligent causes can account for such ‘specified information’. In fact, Meyer writes that intelligent design stands as “the only known cause of specified information-rich systems and, therefore, that ID provides the best, most causally adequate explanation for the origin of the information necessary to produce the first life” (330). I’ll say that I’m not convinced that the overall argument works, but I do find the discussion very interesting.
(Btw, I wonder if appealing so much to ‘information’ and ‘specified information’ and ‘specified complexity’ might distract. Why not just say, “THAT system, THAT process we see in the cell, THAT requires an explanation, no nonintelligent cause can account for those systems, and we know that intelligent causes can.” This might help the ID argument avoid some of the criticisms raised by Jeff Shallit. However, I suppose we would want to specify what property of those systems needed explaining, and we might get sucked back into appealing to specified information again. But isn’t it clear that some property X — even though we’re not sure how to specify X — of those systems needs explaining? Yes. And even though we don’t know how to carefully define X, isn’t it clear that the best explanations appealing to nonintelligent causes are failures in explaining the instantiation of X, and intelligence is known to be a causally adequate explanation of other instantiations of X? Yes. Again, it’s not obvious to me that we should infer intelligent design (even if we answer all of my rhetorical questions with ‘yes’), but I wonder whether it’s necessary to specify that property so long as we have a solid intuitive grasp.)
ALL THAT said, I was wondering what people thought about this little thought experiment Meyer uses.

Imagine a team of researchers who set out to explore a string of remote islands near Antarctica. After many days at sea, they arrive on an icy windswept shore. Shouldering their packs, the team hikes inland and eventually takes shelter from the bitter cold in a cave. There, by the light of a small campfire built to cook their freeze-dried rations, they notice a curious series of wedgelike markings vaguely reminiscent of Sumerian cuneiform. It occurs to them that perhaps these scratches in the rock constitute some sort of written language, but dating techniques reveal that the markings are far more than five hundred thousand years old, far older than any known human writing and, indeed, far older than anatomically modern human beings.
The researchers investigate other possibilities. Perhaps the markings are animal scratchings. Perhaps they were left by some sort of leeching process or by glacial action, perhaps in conjunction with winds bringing sand through gaps in ice at high speeds. After extensive research by investigators with a broad range of expertise, these and other explanations invoking purely mindless undirected causes fail to explain the evidence. An additional discovery reinforces this conclusion. In a broad cavern farther inside the cave, the explorers find a series of drawings on the wall of various fish, birds, and mammals… In the process of their painstaking investigation, the explorers make an inference. They note that, although the markings do not reveal the identity of the scribes, they do point to intelligent activity of some kind. (pp. 373-374)

In these cases, is it plausible to infer intelligence? Let’s stipulate that the investigators know that these markings had to have occurred before human language and drawing abilities developed. Would it still be plausible?

  • Mike Almeida

    In these cases, is it plausible to infer intelligence? Let’s stipulate that the investigators know that these markings had to have occurred before human language and drawing abilities developed. Would it still be plausible?
    Andrew, a quick question. Suppose you find markings in the sand that look just like the English ‘Hello there!’. But unknown to you this is simply the result of objects moved randomly across the sand by the wind. What would you say here:
    1. The markings are (some) evidence of intelligence until you discover that they were made randomly.
    2. The markings are (some) evidence of intelligence even after you discover that they were made randomly.
    3. The markings are not evidence of intelligence.

    July 20, 2010 — 16:07
  • Andrew Moon

    Hmm, well, I’m actually an internalist about evidence. (So, instead of saying the markings are evidence, I’d say that my justified/warranted beliefs about the markings are the evidence.) Anyway, I’d go with 2. I may gain a defeater afterward, but the initial evidence still counts as evidence.

    July 20, 2010 — 16:27
  • Mike Almeida

    So, instead of saying the markings are evidence, I’d say that my justified/warranted beliefs about the markings are the evidence
    But isn’t the marking the evidence for the justified belief that is the evidence that the markings are designed? Or is the evidence for the justified belief yet another belief? (may Pappas and Swain forgive me, but this is why I try to avoid epistemology!).
    Anyway, if you go with (2), then you very likely take the markings in your example to be evidential (or your justified beliefs about the markings to be evidential…). They would be so (for you, it seems) even if there were no intelligent life around to make them.

    July 20, 2010 — 17:04
  • ohhhh, you shouldn’t avoid epistemology! that’s where the best stuff is! =)
    I think that there’s a perfectly fine sense of the word ‘evidence’ according to which facts or states of the world count as evidence. I guess I take ‘evidence’ to denote, roughly, that which persons justifiably believe on the basis of. And those things tend to be internal states (or perhaps psychological attitudes such as knowings and seeings and rememberings). So, I believe that orange has vitamin C on the basis of my belief that that scientist on t.v. said so, and so on.
    In this case, I MIGHT justifiably believe that intelligence caused those markings on the basis of my beliefs that those markings exist and that those markings exist only if intelligence caused those markings. I believe that those markings exist on the basis of my being appeared to a certain way, or on the basis of my seeing them.
    But I guess I’m not sure whether I (or those explorers in Meyer’s example) ought to believe that those markings exist only if intelligence caused those markings. I wonder if it’s properly basic.

    July 20, 2010 — 17:21
  • Is the story supposed to imply that the markings have specified complexity? The story as given in the passage is consistent with the hypothesis that the markings are entirely random.

    July 20, 2010 — 19:17
  • Andrew Moon

    yeah, it is. I think that Meyer would think that the markings have specified complexity. Btw, Meyer doesn’t use this illustration to try to prove anything; it’s just an illustration of when it’d be plausible to attribute design. (He uses it to illustrate other points, but those points are beyond the scope of what I want to talk about.)
    One thing I liked about the story was that I thought that all paradigm cases of plausible design inference were cases when we already know that an intelligent mind exists. I can infer that the graffiti on the wall was produced by a mind, but, I thought, only because I already knew that humans exist. But the story illustrates that even when we don’t know that minds exist, or have no independent evidence that minds exist, we can infer that minds exist from certain sorts of effects. I know that this point might be obvious to some, but it wasn’t to me.

    July 20, 2010 — 19:53
  • What specification do you think he means for the markings to satisfy?

    July 20, 2010 — 21:49
  • Andrew Moon

    Honestly, I don’t know. Unlike Dembski (who focuses a lot on specified complexity), Meyer focuses more on what he calls specified information. The idea wasn’t too clear to me (hence my long parenthetical remark in the opening post). Anyway, what got me were the drawings of the animals. Maybe those count as specified.

    July 20, 2010 — 21:55
  • Gregory Lewis

    I won’t attempt to offer any sort of comment on the epistemology as I’m wildly under qualified, but:
    The X in question seems obviously important as to what the ‘problem’ is here. I haven’t seen any use of Complex Specified Information/Specified Complexity or similar outside ID, and from my limited exposure, I can’t make these ideas make sense (and, given Shallit etc. I’m not the only one). Put archly, I doubt whether there’s anything *to* understand about these things because I think they’re conceptually bankrupt.
    (Put it another way: suppose some fringe group of philosophers invents some addition to modal logic that shows something really controversial like God doesn’t exist. Yet this ‘modality+’ isn’t used anywhere in the field except by those controversially pushing this particular argument – further, all the people who do know a fair bit about modal logic criticize ‘modality+’ fiercely. When you read about ‘modality+’ by its proponents, you struggle to make any sense of it. Is it really worth straining to see if you can divine some sensible view out of ‘modality+’ – why not just dump it?)
    But that isn’t so important – it might be hard to articulate what the X is, and the current attempts (CSI/SC) fail, but maybe there’s some difficult-to-explain X about (for example) DNA replication. But, speaking as an biologist, I simply don’t see what the candidate X would be to motivate a beefy design inference. If ID proponents can’t provide some sort of principled distinction of how these things ‘can’t happen’ under unguided evolution, then we seem to be left with either a noseeum (‘I can’t see how DNA replication could come about by chance, so it probably didn’t’) or an appeal to intuition (‘DNA replication seems so clever and well organized that intelligent agency must have made it’).
    I don’t think either can be trusted on reflection. Noseeums about biology, or intuitions about what systems seem ‘too elegant’ for evolution to ever explain have a notoriously bad track record (cf. Behe with Complement and Flagella, the Eye, etc.), and biologists should point to even more complicated and elegant systems (like immunity) and lay out the incremental evolutionary story – the progression of this sort of work, by contrast, has an exceptional track record. So, even if they don’t have such a good story for DNA replication yet, they have every prospect of filling in the picture. So why use the ID explanatory stopper?

    July 20, 2010 — 22:32
  • Andrew Moon

    Hi Gregory,
    Thanks for the comment. I’ll pick on this sentence:
    “If ID proponents can’t provide some sort of principled distinction of how these things ‘can’t happen’ under unguided evolution, then we seem to be left with either a noseeum (‘I can’t see how DNA replication could come about by chance, so it probably didn’t’) or an appeal to intuition (‘DNA replication seems so clever and well organized that intelligent agency must have made it’).”
    Well, Meyer’s whole point in appealing to the origin of life is that you cannot appeal to unguided evolution to explain the origin of life. Evolution works on pre-existent life, or at least something that can already self-replicate. It doesn’t explain the origin of these systems of self-replicating molecules. So, Meyer does have arguments for why DNA replication couldn’t come about by unguided evolution. Btw, he also has arguments for why they couldn’t come about by chance either. (He takes a couple of chapters on this; note that the appeal to chance dropped out by the early 1970’s).
    Btw, I would want to go back and see what Meyer himself meant by ‘specified information’; I read that chapter a while back.
    Back to what I called ‘X’ in my opening post, I guess I still think that the whole DNA replication process cries out for explanation, even if I can’t explain what about it produces in me the intuition that explanation is needed. And all of Meyer’s arguments that unintelligent causes can’t produce ‘specified information’ also show that unintelligent causes couldn’t produce that process. And we know that minds are the sort of entities that could create such processes. That’s what Meyer needs for those crucial premises in his argument. So, I’m not sure if specifying X is needed.
    Or maybe only a partial explanation/definition of X would be needed. I’m still thinking out loud.

    July 21, 2010 — 0:13
  • Gregory:
    I wouldn’t say that there is nothing to CSI. As I understand it, CSI comes from the fact that there are three somewhat natural probability measures on physical arrangements. For definiteness, think of physical arrangements as black-and-white pixel patterns on a screen, and then there are 2^n arrangements where n is the number of pixels.
    There are three different fairly natural probability measures on this.
    1. There is what one might call “a rearrangement or Humean measure” which assigns every arrangement equal probability.
    2. There is “a nomic measure”. Basically, the probability of an arrangement is the probability that, given the laws (and initial conditions? we’re going to have two ways of doing it–one allowing the initial conditions to vary, and one to vary), such an arrangement would arise.
    3. There is what one might call “a description measure”. This is relative to a language L that can describe pixel arrangements. One way to generate a description measure is to begin by generating random finite-length strings of symbols from L supplemented with an “end of sentence” marker which, when generated, ends a string. Thus, the probability of any string of length k is m^-(k+1) where m is the number of symbols in L. Take this probability measure and condition on (a) the string being grammatical and (b) describing a unique arrangement. The resulting conditional probability measure on the sentences of L that describe a unique arrangement then gives rise to a probability measure on the arrangements themselves: the description probability of an arrangement A is the (conditionalized as before) probability that a sentence of L describes A.
    Basically we have the less anthropocentric nomic and rearrangement measures, and the more anthropocentric description measure. The rearrangement measure has no biases. The nomic measure has a bias in favor of what the laws can produce. The description measure has a bias in favor of what can be more briefly described.
    We can now define CSI of two sorts. An arrangement has specified rearrangement (respectively, nomic) complexity, relative to a language L, provided that its arrangement (respectively, nomic) measure is much smaller than its L-description measure. (There is some technical stuff to be done to extend this to less specific arrangements–the above works only for fully determinate arrangements.)
    For instance, consider the arrangement where all the pixels are black. In a language L based on FOL, there are some very short descriptions of this: “(x)(Bx)”. So, the description measure of the all-black arrangement will be much bigger than the description measure of something messy that needs a description like “Bx1&Bx2;&Wx3&…”. On the other hand, the rearrangement measure of the all-black arrangement is the same as that of any other arrangement. In this case, then, the L-description measure of the all-black arrangement will be much greater than its rearrangement measure, and so we will have specified rearrangement complexity, relative to L. Whether we will have nomic rearrangement complexity depends on the physics involved in the arrangement.
    All of the above seems pretty rigorous, or capable of being made so. So insofar as this is CSI (and while the details are not the same as Dembski’s, it’s a pretty obvious way to develop his approach), I don’t think it’s correct to say that there is nothing to it.
    Now, given the above, we have the philosophical question: Does CSI give one reason to suppose agency?
    There, we have an initial problem. The concept of CSI is language-relative. For any arrangement A, there is a language L1 relative to which A lacks complexity and a language L2 relative to which A has complexity. So CSI had better be defined in terms of a privileged kind of language. I think this is a serious problem for the whole approach, but I do not know that it is insuperable. For instance, easily inter-translatable languages are probably going to give rise to similar orders of magnitude within the description measures. We might require that the language L be the language of a completed and well-developed physics. Or we might stipulate L to be some extension of FOL with the predicates corresponding to the perfectly normal properties. There are tough technical problems here, and I wish Dembski would do more here. Call any language that works well here “canonical”.
    Once we have this settled, we can now ask: Is there any reason to think that CSI is a mark of design?
    Here, I think Dembski’s intuition is something like this: Suppose I know nothing of an agent’s ends. What can I say about the agent’s intentions? Well, an agent’s space of thoughts is going to be approximately similar to a canonical language (maybe in some cases it will constitute a canonical language). Without any information on the agent’s ends, it is reasonable to estimate the probabilities of an agent having a particular intention in terms of the description measure relative to a canonical language.
    But if this is right, then the approach has some hope of working, doesn’t it? For suppose you have nomic specified complexity of an arrangement A relative to a canonical language. Then P(A|no agency) will be much smaller than the description measure of L, which is an approximation to P(A|agency) with no information about the sort of agency going on. Therefore, A incrementally confirms the agency hypothesis. The rest is a question of priors (which Dembski skirts by using absolute probability bounds).
    I think there are two serious problems:
    1. The question of canonical languages.
    2. We do have some information on the ends of agents in general–agents pursue what they take to be valuable. And the description measure does not take value into account.

    July 21, 2010 — 9:17
  • Gregory Lewis

    I haven’t read Signature in the Cell, so I’m going off the secondary literature and what you’ve said. Apologies in advance if I misrepresent anything.
    However, if Meyer really uses the machinery of DNA replication as a means of showing how you couldn’t get a first replicator, that’s just completely mistaken. No modern theory of abiogenesis I’m aware of thinks that the first replicators/protocells popped into being with the DNA replication machinery (or even DNA at all). Rather the whole DNA replication machinery evolved from some other much simpler protocell (perhaps running directly off RNA or other nucleotide derivatives).
    So if Meyer’s argument is something like “There’s no good explanation for abiogenesis (getting our ‘first life’ or replicator) because it would need DNA replication and there’s no way the relevant processes could throw that up”, that is far from good enough – the field doesn’t think the first life had (or would need to have) DNA or DNA replication. If Meyer engages with these theories of abiogenesis and has some principled reason as to why you couldn’t ever get a first replicator by any means, fair enough (though, as we aren’t sure what the ‘first cell’ was, pardon my incredulity). If instead Meyer says that DNA replication is uber-complicated and there’s a reason to think it couldn’t arise by unguided processes like evolution, then I think my prior criticism stands. I could be missing something.
    I can just about follow the contours of what you said, although I won’t pretend to have the mathematical facility to evaluate the detail.
    Let’s grant that Prussian CSI can overcome the conceptual difficulties you note and become something substantive. My main worry is I don’t see how you could deploy this in biological systems to get design. Dembski’s work on this, if I recall relies on Behe’s very dubious stuff on irreducible complexity (as, without it, I think Dembski would need to consider ‘routes’ through the relevant probability space – so P(protoprotoflagellum)*P(protoflagellum|protoprotoflagellum), and so on. and I can’t see any way for him to evaluate these).
    I’ll offer further commentary if I’m convinced that what I’ll say will be more than a parade of my own mathematical incompetence. 😉

    July 21, 2010 — 14:19
  • Gregory:
    I have no idea how to calculate the nomic probabilities in the biological case. However the challenge of calculating the nomic probabilities is just as pressing for the orthodox Darwinian as for Dembski. Unless we actually have good lower bounds on the nomic probabilities for something like a flagellum, we really are not in a position to say that the Darwinian account provides a satisfactory alternative to design.

    July 21, 2010 — 14:44
  • Andrew Moon

    No problem w/not having read the book (it’s a long book!).
    “No modern theory of abiogenesis I’m aware of thinks that the first replicators/protocells popped into being with the DNA replication machinery (or even DNA at all). Rather the whole DNA replication machinery evolved from some other much simpler protocell (perhaps running directly off RNA or other nucleotide derivatives).”
    Yeah, Meyer’s aware of that and spends a chapter discussing the RNA world hypothesis (as I understand it, the hypothesis that the DNA replication machinery ultimately arose from RNA strands that could self-replicate), and he gives about 5 or so arguments why it won’t work. His most notable argument aims to show that many of the same problems for theories that attempt to explain the existence of self-replicating DNA are also problems that attempt to explain the existence of the simpler self-replicating RNA. (It does seem that the problems with the latter are smaller; still, the problems seem daunting and, in my opinion, insurmountable.) Btw, I thought that this was one of the few chapters of the book that contained original arguments with very recent data, discussing scientific publications from the last couple of years.
    (As far as I know, the only theory that explains the origin of the DNA replication system from a prior self-replicating molecule is the RNA-world hypothesis. Is that right? Appeals to protein-first or DNA-first theories, whether by chance or necessity (or both), do not involve evolution. What else other than a self-replicating RNA molecule could possibly do the job? I’m wondering if you know, as a biologist.)
    Of course, much of the discussion in the RNA-world chapter was way beyond my biology abilities, although my bio-friends seemed pretty convinced. Also, I don’t know if his five (or so) arguments there are robust enough to survive what’s on this blog post:
    which cites a very recent article. I skimmed the post and decided I might read it some time in the future. Anyway, thanks for the comments; it’s nice to discuss these ideas w/you.

    July 21, 2010 — 15:06
  • Gregory Lewis

    Couldn’t evolutionary biologists just point to how the relevant bits of natural history and cell biology are suggestive of this ‘Darwinist route’, even if the probabilities are inscrutable? It would hinge on something like these sorts of appearances (like the circumstantial evidence supporting the evolutionary path the flagellum took) confirms Orthadox Darwinism over ID, because it’s a bit surprising the designer would design to give the misleading impression of non-design.
    Fair enough Meyer talks about RNA-world. I’m hesitant to criticize a book I haven’t read (although the sciency blogosphere has trashed it). I do find the focus on the DNA replication machinery a bit bizarre: if you want to focus on origin of life being implausible, you should really just engage with actual origins of life and show how there’s no hope of any of them working. The DNA replication process (which is pretty unanimously though to have come later) doesn’t seem relevant.
    The reason RNA is much ‘easier’ than DNA is because it can do catalysis and coding. DNA doesn’t seem to have much in the way of catalyic properties, and there’s no easy way that one protein sequence can ‘template’ another. That RNA can do both of these things makes the idea of self-replicating or sort-of self-replicating RNA molecules more palatable (especially as we know the RNA can do the sort of catalytic things we want, like stick amino acids together or cleave/add RNA strands). It seems miles less problematic by suggesting how you’d pop in DNA and the proteins needed to replicate it. Maybe that’s insurmountably hard, but there’s a lot of literature to engage with here.
    (This isn’t my field and it’s been a long time since I looked, but the tentative consensus seems to be something like the RNA world: some people think that actually the very earliest organisms used some RNA-like thing which ‘modern’ RNA is a successor. I’m unaware of any protein-first or DNA-first models: as I’ve said, both seem hard because DNA can’t do much, and proteins can’t replicate much. There are issues as to whether this ‘primordial genetics’ happened before a primordial metabolism or a primordial cell membrane, but they don’t seem directly relevant here.)
    The Yarus et al work that Hunt is talking about seems to be filling in one of the ‘next steps’ of the RNA story – getting RNA coding for sequences of amino acids, instead of just templates for itself. If one of Meyer’s ‘insurmountable problems’ is how you’d establish arbitrary codon assignments for given amino acids, then this sort of work seems to neatly answer that problem – that the codon/anticodons preferentially bind the amino acid they code for is suggestive that there wasn’t an arbitrary assignment of codons, but rather some sort of selection done on affinity.
    I think I’ll try and read the Chapter in Meyer’s book. As you say this is citing recent work and offering original arguments, it is perhaps the best the book will have to offer. If I’m impressed I might well read the rest. If not, I’ll try and articulate why: because there’s a remarkable difference in how ID stuff is treated by philosophers and how it is treated by biologists. One of these groups is surely ‘missing something’ – and, ‘cos I’m a biologist, it’s obviously you guys. 😉

    July 22, 2010 — 7:06
  • Gregory:
    Yes, one can point to evidence that the route was taken. But there is a further question, essential to the confirmation of orthodox Darwinism over interventionist ID, whether the route was probable. Suppose I know all the steps in the evolutionary history of the flagellum, and they are just like orthodox Darwinism says they are, but I don’t know whether the steps are probable. Let’s say one step is: “At t4, mutation m10 occurs.” If it turns out that this step is in fact highly improbable, and that the sum total of the probabilities over alternate routes is also low, then the interventionist ID advocate can say: “I grant that at t4, mutation m10 occurred. But God’s miraculous intervention was the cause of m10 occurring at t4.”
    Now, maybe, the orthodox Darwinian can point to evidence that m10 didn’t happen through miraculous intervention. But we do not at present, and are not likely in the future, to have much information on how exactly particular mutations hundreds of millions of years ago occurred. Did a cosmic ray hit a DNA molecule? Was there some chemical damage? Etc. Let’s say that we, through an amazing scientific feat, identified it as a cosmic ray and identified which astronomical object it originated from. If the probability of the cosmic ray being aimed precisely so as to cause m10 at t4 was low (as it obviously was), and if the sum of the probabilities over alternate routes was equally low, the interventionist ID advocate could still insist that God aimed the cosmic ray just so.
    So it still does seem to me that the probabilities are essential to the confirmation of orthodox Darwinism over interventionist ID.

    July 22, 2010 — 11:43
  • Andrew Moon

    Hi Gregory,
    Thanks for the informative response. The focus on DNA replication was more my focus, not Meyer’s. Meyer focuses on ‘specified information’ in DNA, but I wanted to avoid that talk because I wasn’t sure what it meant. I just took another look at the chapter where he defines it (chapter 4, especially p. 91), and he seems to be defining it as ‘information that performs a function’. I have a reasonably firm grasp of what information is and what functions are, and so I think I have a reasonably good grasp of what specified information is. And the origin of specified information in RNA, Meyer thinks, is a significant problem.
    Here’re the relevant sections from the RNA-world chapter that moved me to think it’s not likely to work. (There were a lot of criticisms he made that I didn’t understand, indeed, I am not sure I understand exactly how the RNA-world model works. But these quotes I did understand, for the most part.)
    He writes, “the first self-replicating RNA molecules themselves would have needed to be sequence-specific in order to perform the function of replication, which is a prerequisite of both natural selection and any further evolution toward cellular complexity…” (313)
    After saying a bunch on why it is extremely improbable that such RNA strands arose by chance, he adds, “To make matters worse, as Gerald Joyce and Leslie Orgel note, for a single-stranded RNA catalyst to produce an RNA identical to itself (i.e., to “self-replicate”), it must find an appropriate RNA molecule nearby to function as a template, since a single-stranded RNA cannot function as both replicase and template…
    Even if an RNA sequence could acquire the replicase function by chance, it could perform that function only if another RNA molecule–one with a highly specific relative to the original–arose close by… Orgel and Joyce have calculated that to have a reasonable chance of finding two such complementary RNA molecules of a length sufficient to perform catalytic functions would require an RNA library of some 10 to the 48 RNA molecules.” (315)
    Yeah, I would recommend the RNA-world chapter, chapter 14. I might also recommend chapter 4, the chapter on information.
    Also, a number of scientists have endorsed Meyer’s book: Philip Skell (National Academy of Sciences, Penn State), Scott Turner (environmental and forest biology, State University of New York), John Walton (Organic Chemistry, U. of St. Andrews), Norman Nevin (genetics, Queen’s University, Belfast). These are just some of the scientists who have endorsements written at the back of the book. Dean Kenyon helped him write the book (see acknowledgments section), and I’ve learned since reading the book that Kenyon’s theory of the origin of life was the standard for a time in the 70’s.
    Haha, of course an argument for intelligent design would take some degree of both scientific and philosophical expertise! And, of course, a number of philosophers are very unhappy with Meyer’s book.

    July 22, 2010 — 13:21
  • Gregory Lewis

    That seems right to me, but the prospects for getting any of the probabilities seems inscrutable (I have no idea what the values would be). Worse, even if we could calculate it for the given route, I really have no idea to calculate all other possible routes (especially as our ‘end points’ may well be functional: we need to consider other structures that have never evolved at all!) Now we can expect (I think) the actual route to be from the ‘most probable end’ or the available routes, yet that still isn’t much to go on. Epistemic probability wise, I suspect there will be on reflection a very large number of routes each with a very low probability.
    What I was suggesting is if that really is inscrutable (so there won’t be some nice ‘mutation by mutation’ account by evolution, and all ID ‘you can’t get to there from there!’ arguments fail), the circumstantial evidence confirms evolution over ID. Simplest example would be the phylogenetic tree a nice branching pattern of common descent is exactly what you’d expect if evolution was true (if you see bunnies in the precambrian, so much for evolution!) yet it seems at best unsurprising given ID. So this, I was thinking, confirms evo over ID.
    Thanks for this. I’ll look at Chapters 4 and 14, and come back if I have anything more substantive to add. Two minor notes, though:
    1) I’m not doubting there are some scientists who like Meyer’s book. However the prevailing mood is antipathy – all the bio-bods I know either simply don’t know/ignore it, or have bad things to say about it. I guess the science blogosphere neatly samples this. For the sake of an example, consider ‘The God Delusion’: I’m pretty sure there are a couple of philosophers who’ve made pleasant noises about it, but most either don’t take it seriously or think it’s rubbish.
    2) There’s a nice new book out by Nick Lane on how evolution/darwinism/chemical stuff can get you all the cool cell biology (and other things besides). It’s called “Life Ascending: The ten great inventions of evolution”. I (again) haven’t read it myself (this is turning into a pattern…) but I’ve heard good things about it, so it might be worth a look.

    July 27, 2010 — 7:19
  • Andrew Moon

    Hi Gregory,
    The Dawkins example puts things in perspective. =) However, I will note that although there are some scientists who are experts in the topic of origin of life who would endorse Meyer’s book (e.g., very likely, Dean Kenyon), I don’t know of ANY philosopher of religion who would endorse Dawkins’ book. Anyway, I get your overall point.
    On the second point, thanks for the recommendation.

    August 3, 2010 — 22:51
  • Jeremy Pierce

    there’s a remarkable difference in how ID stuff is treated by philosophers and how it is treated by biologists. One of these groups is surely ‘missing something’ – and, ‘cos I’m a biologist, it’s obviously you guys. 😉
    For all I know, philosophers who comment on these things make all manner of mistakes because of their ignorance of or false assumptions about biology.
    Nevertheless, I can quite confidently state that biologists and philosophers alike who critique ID are especially prone to misrepresent what the ID people are saying, to take them to be arguing for things they don’t even believe and to present their arguments as being other than what they are actually arguing.
    Here are a couple examples:
    I regularly see people equating ID with creationism. Some ID adherents (e.g. Flew) do not even believe in a good God. Others (e.g. Behe) accept the evolutionary account of common descent with animals but insist that an intelligent mind would have needed to guide it. Still others (e.g. Dembski, ) deny common descent but present arguments compatible with it and point out explicitly that they are doing so.
    That means they’re not denying the evolutionary account as a matter of how things took place. They’re just questioning its sufficiency in explaining why it took place. In the same vein, critics among philosophers and biologists both take ID to be arguing for a miraculous intervention of the sort that requires God to be violating the laws of nature or suspending or changing them, but that’s not so for all ID arguments. Some of them are compatible with (and some of the proponents actually assume) a set of laws from the outset that determine the course of origins, but those laws would simply be unlikely without a mind putting them into place. Those arguments (and those aren’t the only kind, but virtually every ID argument can be framed this way, and most are) are thus not necessarily arguing for efficient causes to be any different from what biologists claim. They just think there needs to be a final cause, a purpose of goal, as the best explanation. It’s an inference to the best explanation argument, not an argument that some gap in explanation requires an intelligent mind but an argument that it’s best explained by an intelligent mind. But opponents regularly misrepresent it in this sort of way.
    I’ve also repeatedly seen the claim that the political goals of the Discovery Institute mean that they’re not interested in truth but only in outcome, which is just plain nuts and about as uncharitable a projection as there can be. To claim that they would like people to believe an opinion they happen to think is true somehow means that they don’t really think they have good reasons to think it is simply crazy. But that’s what it amounts to when someone claims that their political goals mean that they are purveying religion as if it’s opposes to free philosophical and scientific inquiry, something that they very much intend to be engaging in, despite claims to the contrary. Having a religious motivation doesn’t mean you have no theoretical basis based on arguments. This fallacious inference worked itself into Wikipedia’s entry on intelligent design, and I couldn’t convince the editors there who had assumed control over the entry to change it. I’ve discussed this point further
    I’ve also seen a renowned philosopher of science either show either (a) complete confusion about what religion is or (b) redefine religion for political purposes. There’s no other way to generate the conclusion that one is being religious merely by believing that there is a supernatural being and thus concluding that all ID is by definition religions. But Elliot Sober does exactly that, and it’s very hard for me to see that as responsible philosophical thinking. I should also say that the Supreme Court has made the same unconscionable mistake, so Sober is in good company, but it doesn’t make it more right that the Supreme Court at one point agreed with him (I have a feeling the current Supreme Court would remedy this if they had the chance).
    I should say that Judge Jones, in the infamous Dover decision, made just about all of the above mistakes and rested his case pretty significantly on some of them, when all he needed to do in his particular case (which is all he was supposed to be deciding) was to explain why the particular curriculum used in Dover was unscientific even by I.D. standards and was based on explicit religious teachings. He instead sought to make a broader claim about ID in general but as a result made amateur fallacious philosophical points into settled precedent in his judicial district.

    August 5, 2010 — 13:08
  • Gregory:
    While the exact values of the probabilities are likely to be inscrutable for a long time, it doesn’t seem absurd that one might eventually get estimates within a couple of orders of magnitude. And that could be good enough to decide the questions.
    “Simplest example would be the phylogenetic tree a nice branching pattern of common descent is exactly what you’d expect if evolution was true (if you see bunnies in the precambrian, so much for evolution!) yet it seems at best unsurprising given ID. So this, I was thinking, confirms evo over ID.”
    One thing the ID folks don’t do enough of is coming up with particular models of how ID is supposed to work answering questions like: Where does agency intervene, in what way and with what motives? (In fact, I think the failure to produce such fairly detailed models is one of the things that distinguishes much of ID stuff from good science.) Now if the ID folks came up with fairly detailed models, your argument might disconfirm some models but not others. For instance, it does (incrementally) disconfirm models that either are incompatible with or unlikely to involve common descent.
    On the other hand, here is a model that your observation does not disconfirm: “There is a God who has certain goals in mind; he would like these goals to be achieved as much as possible by natural causes. He thus sets up an evolutionary environment. Nonetheless, some transitions are extremely unlikely to happen by natural causes. These transitions are the result of divine interventions. These interventions are limited to what is unlikely to happen without intervention, but intervention does indeed happen.” One would, of course, want more detail in this model, but it does seem not unreasonable to think it would generate predictions fairly in line with what we in fact observe. And the hypothesis that God would like his goals to be achieved as much as possible by natural causes is not assumed ad hoc to provide an alternative to evolution: it has theological bases within Western monotheism long preceding Darwin.
    On the evolution side, very early on a common descent model was generally accepted, and this became a part of the particular model of evolution. This was not an inescapable result of accepting gradual development primarily through natural selection. Suppose one day, by a deep-sea vent, we discover non-DNA-based forms of life, and our best science will be that there were two separate abiogeneses: one in the ancestry of the DNA-based organisms and another in the ancestry of those organisms. Both kinds of organism evolved primarily through natural selection, and so on.
    Would we say: “Evolution has been disconfirmed”? I think not. Rather, the dominant model of evolution would have been disconfirmed, but most of what biologists have done would stand unchanged.
    Here are two arguments that my hypothetical discovery would only disconfirm a model of evolution and not evolution itself, or at least nothing central to evolution.
    1. Evolution is often said by biologists to be central to contemporary biology. But if it is central to contemporary biology, then, plausibly, much of what contemporary biologists say would be undermined were evolution disconfirmed. But the hypothetical discovery would disconfirm very little of contemporary biology. So either the hypothesis is compatible with evolution or at least it is compatible with what is central about evolution.
    2. Here is a hypothesis about biologists. Most biologists are quite sure that evolution is true. But I suspect that most biologists do not have the same level of certainty about the claim that there are no non-DNA-based life-forms, resulting from a different abiogenesis, in hard-to-access parts of the earth. Or, to put it differently, most biologists would be very surprised if evolution were disproved, but only somewhat surprised if such life-forms were found. Hence, either their probabilities are inconsistent or evolution does not entail that there was only one abiogenesis in the ancestry of earthly organisms.
    I don’t know that I am convinced by these arguments, but at least the second seems to have some plausibility.
    The point here is that the evidence you give is predicted by a particular model of evolution, not by evolution itself. A particular model of evolution should be compared to a particular model of ID.

    August 5, 2010 — 13:46