<link rel="stylesheet" href="//fonts.googleapis.com/css?family=Roboto%3A300%2C400%2C500%2C700%7CRoboto+Slab%3A400%2C700">Denisovans diverged from humans 4,000 years ago? - Filthy Monkey Men

Todd C. Wood is a geneticist who is also a genuine, dyed in the wool young earth creationist. The co-existence of two such divergant positions is perplexing, made all the more so by the fact he is actually competent: he knows what he’s talking about, yet is still accepts the in-errancy of Genesis. Given the fact he remains relatively scientific, might there be something to this whole creationism kerfaffle?

Wood’s latest “journal” article (I put the journal in quotemarks because it was published in the Journal of Creation Theology and Science) provides a good case study to examine whether he’s onto something. It’s called “Ancient mtDNA Implies a Nonconstant Molecular Clock in the Human Holobaramin.” and is an attempt to calculate when humans, denisovans and neanderthals all diverged from each other by looking at the genomes of those species. Like the man itself, this is competent research with no glaring methodological errors. And yet it still concludes

The calibrated phylogeny places the divergence of Neandertal and ancient Homo sapiens mtDNA before the Flood, about the time of the birth of Noah’s father Lamech

In order to arrive a this conclusion he used phylogenetics, the same process that has revealed we split from neanderthals ~500,000 years ago and from Denisovans ~800,000 years ago. So why did his analysis conclude the split was only ~4,000 years ago? Well, it all comes down to how you build your phylogenetic tree.

When you perform a genetic analysis the only data you get is on how related various creatures are to each other. Whilst interesting, this ultimately doesn’t tell you anything about the “path” of evolution. You can’t work out who the common ancestor was for example, or who descended from who and so forth.  All you’re really left with is a sprawling diagram which looks something like this.

And this isn’t very helpful for working out the family tree of organisms.

To turn this into something which provides more information about evolutionary relationships you have to “root” the tree. This consists of figuring out which of the species your examining is the “outgroup” and then constructing the tree around that. In the above example the outgroup is obviously the mouse, so you “fold” the tree around that.

Yeah, I know that explanation is rubbish but it’s hard to give a better one since the whole process involves complicated calculations and is mostly done by computer. At any rate, the net result of all that fancy mathematics is you turn the above image into this

Much better

Now, that does look a lot more like the phylogenetic trees we know and love but we still haven’t finished. Next we have to work the scale of this tree. You may have noticed the “0.1” line in the above image. How much time does that represent? 1 year? 1 million? 1 billion? Figuring this out is where typical science and T. C. Wood diverge. Normally we like to use some empirical evidence to determine the scale of our phylogenetic tree.

One of the simplest methods for doing so is the “molecular clock” model. Basically, our mtDNA doesn’t get mixed up during sex which means it is only changed by mutations, which are believed to occur at a relatively constant rate. For example, if 1 occurs every year and there are 100 differences between us and chimps then we split 50 years ago (since half the mutations accumulated in each lineage). Of course, this method isn’t perfect and so scientists like to use other techniques as well, although these aren’t particularly relevant to this discussion (and really hard to explain).

So how does Wood work out the age of the human family tree? Is he using some observational evidence he collected from studying known mutation rates etc?

The simplest method of inferring divergence dates would be to calibrate the midpoint root at 6000 years ago

No, he’s basing his genetic scale off the Biblical time scale. As such he can’t really use any of this work to try and provide evidence for creationism since it requires assuming creation to begin with (and in fairness to him, he doesn’t try to). As such the phylogenetic tree he produces is basically useless unless you presume the Bible provides an accurate age of the earth.

Wood’s reconstruction. The blue line represents the flood.

So if you like to base your opinions off observable evidence then you can basically ignore his research and go home. However, before you do so I’d like to mention 4 interesting implications of his work.

  1. You have to reject the various empirical evidence that science typically uses to scale their phylogenetic trees. So right off the bat we’re in the business of evidence denial to make this worldview work.
  2. Even when you do substitute the evidence with your desired assumption you still don’t get the results you want. When Wood rooted the tree 6,000 years ago the results indicated the last common ancestor of all people lived ~1,000 years ago, which is absurd. In order to make his worldview work he had place the common ancestor of the specimens he is studying ~4,000 years ago, but there’s no science behind this decision.
  3. In order for humans, denisovans and neanderthals to have split ~4,000 years ago mutations must’ve been accumulating at a rate over 300 times faster than even the most generous figures obtained by studying modern mutation rates.
  4. Since the divergence was before the flood yet denisovans and neanderthals are still contained within us that means their lineage must’ve survived the flood. Wood suggests that two of the four women on the ark were denisovan and neanderthal.

So at the end of the day there is nothing about Wood’s work that suggests creationism is a viable option. Although he’s a competent researcher his work is hampered by the fact that he is basing his work off the assumptions of his worldview rather than anything observable or demonstrably true.

Related posts



Categories: Creationism

12 Comments

Jim Birch · 4th September 2012 at 4:45 am

It’s funny or terrible but in the end beliefs, even scientific beliefs, are just beliefs and there is a kind of aesthetic judgement involved in which ones you choose. Normally in science the aesthetic is elegance, the desire for economical answers with the minimum complexity required by the data. This a wild, and wildly successful, shift from the narrative approach to knowledge that ruled before science.

However, given sufficient emotional need, we see that someone can believe just about anything at all, no matter how many dissonant consequences it entails. You might think that Wood, as a trained geneticist, might have noticed but he hasn’t, or won’t, or can’t. It’s an illuminating, if disconcerting, window on the limits of human rationality that we content with in others and ourselves. As a confirmed atheist I’m tempted to cynically add “There, but for the Grace of God, go I” but on consideration there are plenty of religious believers who haven’t taken this kind of crazy turn.

    Adam Benton · 5th September 2012 at 11:03 am

    When I was first learning about parsimony my first reaction was “why does that make it more valid.” Just because an idea has fewer assumptions doesn’t make it right it just means more evidence is needed to show the alternatives are true.

      Jim Birch · 6th September 2012 at 8:23 am

      And we don’t, and can’t, actually know what a thing really is, except by guessing. All we can really know is the information we can collect about it from its interactions. This is most obvious in physics where asking what an electron really *is* is crazy; we equate the electron with it’s observable properties.

      In the “real” world of objects and processes that are wildly more complex than an electron, we have the hubris to assume that we actually know what things are. Yet these assumptions of knowledge are typically made from imperfect knowledge of a few gross features and the feeling of knowing we get is often little more than habituation. Parsimony is a strategy that guards against habitual thinking, flights of fantasy, wishful musing and the like. The parsimonious explanation may not be right, but it’s likely a vast improvement to the alternative of just making stuff up.

        Adam Benton · 6th September 2012 at 8:40 pm

        Don’t get me started. When you start to think about how we aren’t actually seeing something, just the photons which bounced off it (and then that is being interpreted via our subconscious) the whole thing starts to get weird and confusing. Whilst it might be tempting to simply ignore these facts, I think they’re important guards against the hubris you mention.

eyeonicr · 10th September 2012 at 9:41 am

What happens if you arbitrarily decide that modern humans (or an amalgamation of all three human groups) are the outgroup? And have any creationists tried it?

    Adam Benton · 10th September 2012 at 5:20 pm

    Honestly, I have no idea. My knowledge of applied genetics is very poor. However most “how to” guides on the subject note that using a species as an outgroup that isn’t really an outgroup is very bad.

    cavernicola · 26th September 2012 at 4:01 am

    It won’t change a lot; the only thing is that you’re gonna have an unrooted tree; so you won’t be able to determine ancestry; but the relationships between the species will be the same. AFAIK you don’t set an outgroup explicitly, you just add something so different that the pyhlogentic software assumes it’s the outgroup.

      Adam Benton · 1st October 2012 at 5:31 pm

      How does using an outgroup “calibrate” the tree and tell you when various splits occurred?

        cavernicola · 2nd October 2012 at 8:19 pm

        The outgroup is a taxon outside, but not too far, from the group you’re studying; it is expected to have ancestral characters (to be plesiomorphic) to your ingroup. All the other subjects of your study must be more closely related to each other than to the outgroup; so the outgroup will branch of first at the root of the tree; it will set the base of that phylogeny. I mean it will say where to start, which one is the first node, and thus give order to the tree but it won’t change what happens inside it.
        In your post in the unrooted tree if you start with mouse an fold the branches you’ll get to the same tree as the rooted one. How does mega/phylip (whatever they used) know that mouse is the outgroup? Because it’s the one with the most ancestral characters and the most divergent (it’s determined in the same way as the other branches, using an algorithm such as NJ, ML, UPGMA, Bayes …).
        I’m not sure if i explained myself (I can blame that my evolution course was 4 years ago and that english is not my 1st language)

Leave your filthy monkey comments here.

This site uses Akismet to reduce spam. Learn how your comment data is processed.