The Language of God: Size Doesn’t Matter

The Language of God, Chapter 5

By B.J. Marshall

In this section, Collins describes how “the study of genomes leads inexorably to the conclusion that we humans share a common ancestor with other living things” (p.133-4). There are a lot of ideas in this chapter to unpack, so I’d like to start by reviewing Collins’ material on DNA at a high level: size and broad similarity. It would be my hope that, even if Creationists only heard some of the material in this chapter, they would quickly see the flaws in Intelligent Design.

First, to lay some groundwork: There is a common misconception made particularly among Creationists that atheism necessarily follows from believing in evolution. This argument is as logically flawed as saying that atheism necessarily follows from believing in gravity, eschewing the entirely-poorly publicized view of Intelligent Falling. Granted, I think evolution poses difficult challenges to theism, but evolution is not theism’s death knell. Atheism is a lack of a belief in god(s); the strongest form of atheism takes it one step further to declare “there are no gods.” That’s basically it. One could hypothetically be an atheist and still believe in the Tooth Fairy, The Loch Ness Monster, or the efficacious treatment of homeopathy. Or, more to the point of evolution, an atheist could believe in Intelligent Design so long as the designing was done intelligently by some space-faring aliens. So long as you don’t believe in a god, you’re an atheist; this is despite all the other nonsense you might believe. Of course, Collins wants you to see that, too. He’s trying to harmonize science and religion, so it would do him no good whatsoever to make all these compelling assertions about the validity of evolution if he thought you were going to jump ship and become an atheist.

Now, that being said, Collins called the genome “the book written in the DNA language by which God spoke life into being” (p.123). Although we still don’t know how abiogenesis occurred, Collins appears to be incorrect in waxing poetic about DNA’s crucial role. It appears that RNA, acting as simple enzymes, might have paved the way for life to begin. But even beyond waxing poetic, Collins treats DNA as something sacred, as “uncovering this most remarkable of all texts was both a stunning scientific achievement and an occasion of worship” (p.3). And, although Collins has admonished his readers to steer clear of god-of-the-gaps arguments, he states that “DNA… seems an utterly improbable molecule to have ‘just happened’” (p.91). Indeed, Collins confesses that he is “in awe of this molecule” (p.102) and regards the “digital [sic] elegance of DNA” as “deeply satisfying” (p.107). (Since DNA is based on four letters, it’s really a Quaternary system.) It’s probably that deep sense of awe that one can only get through belief in god.

Regarding the size of the genome, Collins makes the observation that a surprisingly small portion of it actually codes for proteins. There are only about 20,000-25,000 protein-coding genes in the human genome. Collins states that the total amount of DNA used by those genes to code for protein is about 1.5-2.0% of the total genome. Collins notes that some observers have been insulted at this. Surely as monarchs of the animal kingdom, we should be special! Well, these observers contend, perhaps “our complexity arises not from the number of separate instruction packets, but from the way they are utilized. Perhaps our component parts have learned how to multitask” (p.125). He disappointingly never expounds on whether genes can multitask, so we’re left wondering. Answer: at least some can.

To quell the righteous indignation of some, Collins offers an analogy by way of the language used to write books. He goes on to say the average educated English speaker has a vocabulary of about 20,000 words. (Actually, that is a very low estimate, as Stephen Pinker in “The Language Instinct” points out that the average high school graduate knows about 45,000 words; it might even be 60,000 if you count proper names.) Collins says these words can be used in simple ways (owner’s manual) or really complex ways (James Joyce’s Ulysses). Unfortunately, I think the analogy fails in a way that subverts Collins’ intent. If you want to bring the faithful around to seeing how evolution (unplanned, no ultimate goal, no creator) works, then it doesn’t do well to relate it to a construct (language) whose entire function hinges on the intent of, and usage by, intelligent actors. I think I would have offered the reader an analogy of chemistry: you got some protons, neutrons, and electrons to form a few basic building blocks. From different combinations of these, you get the periodic table. For the theoretical physicists in you all, you could even punt to the different vibrations of strings a la Superstring Theories.

Regarding the similarities, Collins mentions that humans are all about 99.9% the same DNA-wise. This certainly makes me feel better, since my hometown was jokingly known for its residents meeting their future spouses at family reunions. He points out that this fits well with the fossil record, which places us in East Africa about 100,000-150,000 years ago to a common set of founders about 10,000 in number. Now, before you get all antsy, making your way to AnswersInGenesis (I can’t bring myself to link to it) to refute the fossil record, I want to tell you we will disregard for now this line of evidence; we’re really interested in seeing how Collins can back up his claim that DNA alone is sufficient to demonstrate evolution.

Back to just DNA. Here’s a nifty table!

Animal Protein Coding Genes Random Segments between Genes
Chimpanzee 100% 98%
Dog 99% 52%
Mouse 99% 40%
Chicken 75% 4%
Fruit Fly 60% ~0%
Roundworm 35% ~0%

The significance of that table is this: If you asked a computer to construct a tree of life based solely on similarities of DNA sequences of multiple organisms, you’d get (courtesy of this site):

Tree of Life, made only through DNA evidence

The DNA similarities also show that genetic mutations that do not have deleterious effects on survival will accumulate over time – the stuff quite arrogantly dubbed “junk DNA.” Indeed, mutations in coding regions of genes are observed far less frequently, since the deleterious effects of these are more pronounced.

If these genomes, Collins asks, were created by some intelligent designer, why would these particular features appear? Collins poses more challenges to Creationists, and we’ll address them in the next posts.

Other posts in this series:

About Adam Lee

Adam Lee is an atheist writer and speaker living in New York City. His new novel, Broken Ring, is available in paperback and e-book. Read his full bio, or follow him on Twitter.

  • Chet

    DNA is certainly quarternary, but that doesn’t mean it’s not digital – “digital” simply means an encoding that relies on discreet values, as opposed to analog, where values are continuous within some range. Binary is a form of digital encoding but it’s not the only form.

  • jane hay

    I haven’t read Collins’ book, but it seems he left out the huge issue of the action of genes on each other and on embryonic developmental processes, which is where a lot of that 1% difference between chimps and humans comes in(and where my email handle comes from – evodevo). You are not just the sum total of coded-for proteins. Sean Carroll is a better source of info on this fascinating facet of evolution.

  • AnonaMiss

    As a former student of linguistics, I was a little bothered by your treatment of the language analogy. I’m probably going way over the top and I know I’m being very tangential so I encourage anyone who’s not interested in the subject to just ignore this comment.

    Calling natural languages “constructs” implies a lot of misconceptions about language, most notably the idea that someone constructed them. In fact, the terms that you use to differentiate language from evolution apply very well to natural languages: they are unplanned, they have no creator. As for an ultimate goal, I’m not exactly sure what you mean by that. You can’t mean that languages themselves have goals, because they’re not actors. If you mean that languages are designed with a goal in mind, that’s silly, because natural languages are not designed. If you mean that when language is used, the people using it generally have a goal in mind, that’s a fair contrast.

    Living in a literate society, we often forget that written language is not natural, and ascribe properties peculiar to written language to language as a whole. Written language is planned, has a creator, and sometimes even has an agenda – for example, many written languages were invented for the purpose of converting speakers of the language to Christianity. But written language is a support and a shackle for language, not language itself.

    Preliterate languages change dramatically within a matter of generations, not because anyone wants it to change, but because of the human instinct for slang. With no codified standard to keep close to, vocabulary and even grammar is free to contort and mutate. Language evolution is actually less directed than biological evolution: there are slightly stricter parameters for what can be born, but there is no such thing as a “less fit” natural language, only less powerful communities of speakers.

  • jack

    “DNA… seems an utterly improbable molecule to have ‘just happened’”

    Yes, as you point out, Collins is here lapsing into the god-of-the-gaps fallacy. The answer for this one, of course, is the same as for all the other ‘utterly improbable’ aspects of living things: DNA did not ‘just happen’. It is itself the product of evolution. Its immediate evolutionary precursor, as you ably point out, was almost certainly RNA. But RNA was also, almost certainly, the product of a great many generations of molecular evolution. We don’t know what was the first self-replicating and evolving chemical system on earth. Perhaps we never will. But whatever it was, it was something much simpler than RNA — so simple that its spontaneous formation from its primordial environment was not a matter of utter improbability or divine intervention.

  • DSimon

    [...]there is no such thing as a “less fit” natural language, only less powerful communities of speakers.

    AnonaMiss, isn’t that the same thing, though? Imagine a preliterate community of people who speak a common language. They are split into two groups due to a natural disaster. At that point, the language of one group develops in one direction, and the language of another develops in the other. Group A’s language develops some new features that makes it more useful for that group’s survival; Group B’s unfortunately develops some defects. Isn’t Group A more likely to go on to be the root of further languages, and end up having more influence on the future of overall human thought?

    I’m not sure how much of an effect this process would have on language development, but it seems like it would have to be greater than zero.

  • BJ

    @AnonaMiss: I was referring to the way language is used in various ways, like writing an owner’s manual or Ulysses. I’ve listened to the A Way with Words podcast for awhile now and appreciate the way languages “evolve.” Thank you for expounding on the point and for catching that.

  • CR

    But atheism does necessarily follow from evolution.

    Historically, Believers got around the whole “sickness and dying thing” by blaming it on the Devil. But genetic disease, that is a new sort of animal, because evolution, when understood properly, explains where this comes from. If God exists, and evolution is true, it comes from God.

    Try for a moment to put yourself in the mind of a Believer. If evolution is true, then God is directly responsible for child Tay-Sachs and a host of other horrible things. Talk about cognitive dissonance.

    There are only two ways out. You can either punt to Mystery. God is mysterious. His ways are not our ways. Or you can bite the bullet and admit that if God does exist, then he is nothing like the sort of God you thought you knew. Either he is evil, or he is indifferent to our suffering, or he is a bumbling sort of deity who doesn’t much know what he’s doing. Atheism, or “a lack of belief in the God the Christians keep going on about”, follows readily from here.

  • jack

    Atheism, or “a lack of belief in the God the Christians keep going on about”, follows readily from here.

    “Readily”, yes. But that’s not the same as “necessarily”. As you point out, believers can punt to Mystery (or other theodicies), and a great many of them do. It’s amazing but true that some people spend their entire lives doing these mental gymnastics just to avoid abandoning their imaginary relationship with a loving god. Collins evidently is one of them.

  • Samuel

    “I haven’t read Collins’ book, but it seems he left out the huge issue of the action of genes on each other and on embryonic developmental processes, which is where a lot of that 1% difference between chimps and humans comes in(and where my email handle comes from – evodevo). You are not just the sum total of coded-for proteins. Sean Carroll is a better source of info on this fascinating facet of evolution.”

    You mean the homeobox. It is a set of genes that deal with development. You are the sum of coded protein- what matters is how much and when.

    “As for an ultimate goal, I’m not exactly sure what you mean by that. You can’t mean that languages themselves have goals, because they’re not actors. If you mean that languages are designed with a goal in mind, that’s silly, because natural languages are not designed. ”

    Isn’t the idea of grammer genetic? Pidgin languages don’t have them but children who speak them develop a grammer for them.

    “Group A’s language develops some new features that makes it more useful for that group’s survival; Group B’s unfortunately develops some defects. Isn’t Group A more likely to go on to be the root of further languages, and end up having more influence on the future of overall human thought?”

    Unlikely. Language is very malable. If there is something missing, speakers will borrow it from other languages. Spread of lanuages rely more upon migration and conquest.

  • Reginald Selkirk

    the stuff quite arrogantly dubbed “junk DNA.”

    Sometimes arrogance is justified. Biochemist and textbook author Larry Moran at Sandwalk has run a lengthy set of posts about junk DNA and its misconstrual by Creationists.

  • Reginald Selkirk

    jack #4: But RNA was also, almost certainly, the product of a great many generations of molecular evolution.

    You can lower your degree of certainty about that.

    I’ve heard this one often, and I don’t buy it. RNA has the right combination of properties to be the original molecule of life: the ability to store and replace information, and the ability to catalyse chemical reactions. RNA’s lineage is thoroughly embedded throughout the metabolism of current forms of life, including information storage (some viral genomes), information transfer and replication (mRNA and tRNA), the basic currency of energy (ATP), cofactor in many protein enzymes (NAD and others).

    If something preceded RNA, then it must have been thoroughly displaced, which seems unlikely. And the list of candidates which also contain the right properties to be the first molecules of life is pretty short.

    The strongest argument against RNA-first has been the difficulty of RNA synthesis in prebiotic environments (an environment about which our knowledge is vastly imperfect), and a brilliant new pathway for the synthesis of RNA in 2009 cleared that hurdle. See Powner, M. W., Gerland, B. & Sutherland, J. D. Nature 459, 239-242 2009.

  • Reginald Selkirk

    #7 CR: Historically, Believers got around the whole “sickness and dying thing” by blaming it on the Devil.

    I never understood how that argument carried any water. Where did the Devil come from, didn’t God invent him? It’s pretty hard to pass the buck when you’re the Be-all and End-all.

  • Ritchie

    NO! No no no no no…
    When I clicked on the link in Intelligent Falling I was SURE it was a joke. It HAD to be! Surely it was a link to a joke article, a parody, or perhaps a witty cartoon.
    W…T…F…?!?!
    I need a drink.

  • jack

    You can lower your degree of certainty about that.

    Reginald,

    Consider it lowered! Thanks so much for that reference. It was news to me, and exciting news at that. Many questions still remain, of course, and I still think it likely that a simpler self-replicating system preceded RNA, but now I’m a bit less sure.

  • BJ

    @Ritchie: The Onion, who I linked to via “Intelligent Falling,” is a parody. It was a joke, but they wrap it up to sound legit.

  • BJ

    Non-coding DNA, aka “junk DNA,” is far from junk. It’s shown to play a large part in regulatory effects like turning on and off the expression of genes. Makes me glad to listen to the Skeptic’s Guide to the Universe.

  • Reginald Selkirk

    BJ: Non-coding DNA, aka “junk DNA,” is far from junk.

    What percentage of “junk DNA” has been shown to play a role in regulation? Look at the numbers. Do yourself a favor and search around Larry Moran’s blog for entries on “junk DNA.” Link already provided.

  • Reginald Selkirk

    Specific articles of interest on Junk DNA on Sandwalk:

    More Junk DNA Fallacies
    Most of our genome is junk. That does not mean that all of our genome is junk and it certainly never meant (among intelligent scientists) that all of our non-coding DNA is junk.
    (bold added by me for emphasis – RS)

    I agree with Ryan Gregory that this is extremely misleading. It implies that there are legitimate scientists who think that all non-coding DNA is junk. It would be far better to say something like this …

    Genes that encode proteins, and other genes, make up only a few percent of our genome. If you add in all of the other DNA sequences that are known to be essential you still can only account for no more than 5% of our genome. Most of the rest is thought to be junk DNA with no biological function. There are no respectable scientists who think that none of it will ever be shown to have a function but the general consensus among the defenders of junk DNA is that the vast majority of these DNA sequences, consisting mostly of defective transposons and pseudogenes, will turn out to have no function.

    The authors of the paper go on to present evidence that about 5.4% of non-coding DNA has a function.

    Big deal. That’s not much more than what the textbooks have been saying for several decades.

    Theme: Genomes & Junk DNA
    Junk in Your Genome

    Transposable Elements: (44% junk)1
    DNA transposons: 3%
    retrotransposons: 8%
    L1 LINES: 16%
    other LINES: 4%
    SINES: 13%

    Pseudogenes (from protein-encoding genes): 1.2% junk

    Ribosomal RNA genes: essential 0.05% junk 0.09%

    Protein-encoding genes:
    transcribed region: essential 1.8% junk (not included above) 9.6%
    regulatory sequences: essential 0.6%

    Repetitive DNA
    α satellite DNA (centromeres)
    essential 2.0%
    non-essential 1.0%%
    telomeres
    essential (less than 1000 kb, insignificant)

    Total Essential (so far) 4.5%

    Total Junk (so far) 55%

    1. A small percentage (less than 1%) of all transposable elements have acquired a function in the human genome.


    Genetic Load, Neutral Theory, and Junk DNA

    —————-
    There you go, BJ. The main themes are that “non-coding DNA” and “junk DNA” are not synonymous, and that a large fraction, over half, of our genomes is established to be junk. If you want to persist in your opposition, please bring some numbers and some high quality data to the table.

  • jack

    One more thought on ‘junk’ DNA:

    A long stretch of noncoding DNA that serves no obvious and specific purpose, one that is highly repetitive (and therefore of low information content) or loaded with pseudogenes, may still have a nonspecific but important function: reducing the linkage between genes on either end of it. These stretches of DNA increase the probability of crossover, during meiosis, between the genes they separate, and therefore may alter the dynamics of evolution in subtle but significant ways. Evolution is full of such subtle and nonintuitive tricks, so I think it is best to approach ‘junk’ DNA with some scientific humility. There is much about the genome we do not yet fully understand.

  • BJ Marshall

    Total Essential (so far) 4.5%

    Total Junk (so far) 55%

    I think you meant 45%, not 4.5%, correct?

    I always thought that people meant “non-coding DNA” and “junk DNA” as synonyms. Thanks to your information, I’ve amended my definition. While I understand that parts of that non-coding DNA do have important regulatory functions, I didn’t know just how small that part was – with the rest being actual junk.

    I’ve also subscribed to that blog; it’s really interesting!

  • Reginald Selkirk

    I think you meant 45%, not 4.5%, correct?

    No. That’s a direct cut-and-paste. 55% established junk, <5% essential, that leaves only 40% that could be argued about. The two numbers are not meant to add up to 100%.

    I always thought that people meant “non-coding DNA” and “junk DNA” as synonyms.

    Well, many people do mean them as synonyms, but they are wrong. See Larry Moran’s blog for numerous examples. And press releases using them as synonyms are as common as mainstream media science articles about “the missing link.”

    Happy New Year.

  • jane hay

    Collins also didn’t seem to mention the decade-long research into “small” RNAs and their role in gene regulation. See Science, Vol.330, p.1614 for a brief overview of RNAi and lincRNAs. ~80% of some cells DNA may be transcribed into RNA, but we don’t know exactly what it does. It is NOT translated into protein, however.

  • Ritchie

    @BJ

    Oh right. Whoops. Feeling pretty small now. Time for another drink then… :)