Implications of Intelligent Design

I’ve gotten involved in a fairly lengthy discussion of Intelligent Design on reddit as a result of posting my review of Darwin’s Doubt there. Most of the discussion has centred around a couple of key points, that my interlocutors seem to insist on repeating, ad nauseum.

The first point is the inherent implausibility of Evolutionary Theory with regards to explaining speciation, and the second is their refusal to accept that Intelligent Design is an implicitly religious argument. These two points are critical problems in their treatment of this argument.

Plausibility is relational. The implausibility of an event is entirely irrelevant when determining if it happened, or not. Plausibility matters if and only if we have several competing theories. Given time/resource constraints (and assuming roughly proportional costs), we should start testing the most plausible explanations first. But we can’t simply say “well, since x is more plausible than y, we’ll just assume that y didn’t happen”. That’s not the way we work when genuinely seeking explanations. That’s the way we work when x fits our biases, and we don’t want to entertain that it could be false.

To do some basic math: a human can produce 8.4 million different gametes just by shuffling chromosomes (not even taking into account crossing over) so with two gametes combining it’d be 8.4mil times 8.4mil = 70.6×10^12 possible combinations of DNA from any pairing of any two humans. The odds of the particular set of DNA that I have popping out of my parents is 70.6×10^12 to 1, against. Let me write that out in full: 70,600,000,000,000 to 1. Against. (this information from Freeman, S, Harrington M., Sharp, J. 2014. BIOLOGICAL SCIENCE, 2nd Canadian Edition, courtesy of a friendly biologist)

Let’s tone it down, as I’m sure that not all possible combinations are viable, and just say that there’s a one in a million chance of any particular set of DNA combining from ones parents. Notice that I’ve knocked off 6 zeros there, meaning I’ve eliminated more than 99.99999% of the possible combinations. If more than 0.00001% of the possible combinations of DNA are viable then I am vastly underestimating these odds, which will empower my conclusion.

Now each parent also had a one in a million chance of having the DNA that they do. So the odds of Mum’s DNA (set A) combining with Dad’s DNA (set B), to produce my DNA (set C) would be a million times a million times a million. Go up another generation, and we have 4 more sets to introduce.

We have a mathematical series here, which generalises out to the following: the odds that the particular DNA that I have would appear starting from N generations ago is equal to

(10^6)^(2^N) times (10^6)^(2^[N-1]) times (10^6)^(2^[N-2]) times … times (10^6)^(2^0)

To simplify that down, this is (10^6) to the power of the sum of a sequence, written as follows:

Millions and millions


So k is the number of generations we’re going back, so let’s keep it simple (again) and just go back 10 generations, about 300 years. So if we let k be equal to 10, then our sum becomes 2047, which means that the odds of my particular DNA popping into existence from a mere 10 generations ago is (10^6)^2047, or 10^12282 to one, against.

That bears repeating: the odds of my existence, of my particular combination of DNA, is 10^12282 to one. Yet here I am (and let’s bear in mind that I am underestimating this number by hundreds of orders of magnitude). In order to argue that the author of this particular blog post actually has a genetic sequence of DNA requires far, far more than simply arguing that “that set of DNA is highly implausible”: that argument achieves nothing.

Where plausibility matters is in the comparison of two cases, as mentioned above in the testing example. However, without knowing the possibility of both cases, we can’t simply decide which of the two is more or less plausible. I feel a quote is relevant here:

“How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?” Sherlock Holmes

And that’s the core of it: if one wishes to claim that something was designed, in order to evaluate that claim, we need to know who is claimed as the designer. Logically impossible entities are self-defeating (i.e. self-refuting). Materially improbably entities are, at least, interesting and can’t simply be rejected out of hand (because they’re, y’know, possible). In order for the ID community to be taken seriously, beyond their own community, they need to start fleshing out who or what is the designer.

Meyer, in Darwin’s Doubt, doesn’t do that: instead he calls for the tossing out of Methodological Naturalism. Now… This is kind of a weird response. As I outlined in my review of his book, before we can eliminate methodological naturalism as a method of doing science, we have to demonstrate that all possible natural explanations have failed. But nope, Meyer doesn’t do that, and goes straight to the supernatural.


Well… Intelligent Design has some implications that are likely to discourage his religious (and lucrative) readership. Here’s a diagram.

Possible Origins of Life

Here’s the rub: Meyer (and Intelligent Design proponents in general) want to argue that some non-defined “Intelligent Agent” was involved in speciation on earth. And that’s all fine and dandy, but that doesn’t get you out of methodological naturalism. One of the possibilities that they have to completely eliminate (i.e. demonstrate to be impossible) is alien intervention.

This is simply the logical space in which they are placing their arguments: I am not in favour of the alien hypothesis, nor am I in favour of the god hypothesis. So far, the bulk of the experts (i.e. actual biologists) seem to be under the impression that some variation of Evolutionary Theory is going to resolve this issue in favour of the processes hypothesis. As I am not an expert in Biology, I’m going to defer to their expertise (insofar as their methodology is sound). Simply arguing that the processes option is “implausible” doesn’t allow one to immediately leap to the gods option: that has to be demonstrated to be possible, first, then more plausible than both processes and aliens. Jumping immediately to “a god did it” is a demonstration of the religious nature of ID.

Given that I’ve seen a lot of definitions for “god”, and precisely zero of them have been coherent (that is not self-contradictory), I don’t hold out much hope for the proponents of ID. But I do wish they’d be more honest with their arguments.

[Please note that Mathematics, especially Calculus, is not my strong suit. I welcome anyone pointing out the errors in my math. I suspect, however, that the errors are unlikely to affect the actual core point being made, which is not numerical but logical]

Follow Brian on Twitter!


, , ,

4 responses to “Implications of Intelligent Design”

  1. Hi Brian,
    You should refer your debaters to the 2005 Dover Pennsylvania trial where the (religious) judge ruled that ID(intelligent design) is religiously based and there is no science in it. His ruling was posted on line and his explanation is a good one….I don’t have it handy but I’m sure a google search will turn it up.
    Even if we granted that (a) god created the universe we would still want to know how “god did it”. We have discovered so much of how the universe works naturally that we should be able to see how “god did it” if there is a god. Unfortunately for theists, after hundreds of years, no “hand of god” has been discovered in any and all of the scientific theories we now hold. Every scientific theory we have, including evolutionary theory, is explainable by purely natural means. No god needed. All scientific theories must be able to generate testable hypotheses and independently observable, measurable, empirical results. It doesn’t work simply by trying to discredit competing theories or just claiming a [supposed] better explanation (“god did it”). A successful alternative to evolution would have to not only account for everything evolution already has, but then account for more – like the supposed holes in evolution that ID advocates misguidedly try to poke.
    Evolution is a fact: it happens. Speciation happens. It is also a predictive theory: it explains why things happen and have happened in ways that allow us to find out more about the world.

  2. Hey Randy,

    The issue there is that the judgement is merely that: a judgement. Not only would they (and they do) argue that the judge’s opinion is incorrect, but they also argue that ID has come a long way in the last 10 years. And it has, insofar as its proponents have learned to steer away from the obviously religious arguments, and to not complete arguments that entail religious beliefs.

    Furthermore, the ID folk don’t dispute the fact of speciation (Darwin’s Doubt is entirely centred on that particular phenomenon), they simply dispute the mechanism by which it occurs. And, in fairness, so do many modern Biologists (hence the suggestion for an Extended Synthesis:

  3. Hi Brian,

    The Dover Trial result was more than “merely a judgement”. It was a reasoned and correct evaluation of the evidence presented at trial. There have been other “trials” where the respective judges have ruled against creationism and ID as science. And other than steering away from the religious arguments how has ID progressed? As I said in my previous comment there is no science in ID. It is not a scientific theory. There is no conspiracy to keep scientists from presenting ID as science or presenting it as a valid scientific theory. There are scientists (like those at the Discovery Institute) who would be ever so happy to research ID as a scientific theory but they have nothing to research. So how has it progressed scientifically?

    So what mechanism do the proponents of ID propose that causes speciation? Either some agent “monkeyed” around with ancient ape genes (2001: A Space Odyssey anyone?) and either we can discover this or we can’t. So far we haven’t and with the advances in gene mapping and identification it’s highly doubtful now than we will discover some outside agent causing speciation.
    That leaves natural processes. So what if scientists can’t agree on the mechanism that causes it. That doesn’t mean “god did it” or whatever agent they propose. Quite likely there is more than one mechanism that can start the speciation. Isn’t this what the extended evolutionary synthesis proposes?
    Ask your debaters for evidence that ID is a scientific theory. I’m sure PZ Myers and Gerry Coyne (Why Evolution Is True blog & book) would be glad to see their evidence (as well as many other scientists). As someone who is speaking from experience you need to be wary of arguing
    with um…..IDer’s on their terms. You can’t “win”. And they can’t “win” because they cannot show ID is science as much as they might like to think they can.

  4. The Dover Trial result was more than “merely a judgement”. It was a reasoned and correct evaluation of the evidence presented at trial.

    Yes, I get that it was a judgement that you agreed with. I also think it was the correct judgement, but that doesn’t elevate the status of that particular legal opinion to be more than merely a legal opinion. A different judge in a different time and place? Probably a different opinion, which would be as legally binding as the current one.

    I agree with the rest.

    The argument they have (or at least this particular one has) was an inductive argument at core, and core is unjustified:

    1. If a phenomenon is complex (highly improbable), specified (conforms to a detachable pattern), and irregular (non-redundant), it is designed by an intelligent agent.
    2. Biological systems meet the criteria above.
    3. Therefore, they are designed.

    The issue is with premise one. Where the majority of anti-ID arguers go is to push on the complex, specified, and irregular claims, and query the definitions. This is, frankly, a complete waste of time: quibbling over definitions doesn’t show that the argument is wrong, merely that the definitions are disagreed with.

    The base logical error (and I mean that in the strictest possible terms) is the use of “intelligent agent”. This argument is generalised from the fact that we see design flowing from intelligent-agents-that-are-human, intelligent-agents-that-are-ants, intelligent-agents-that-are-beavers, etc. to just arguing for thus far unobserved intelligent agents. Which, like the “all the swans I’ve seen are white, therefore all swans are white”, suffers from the basic Problem of Induction:

    And in arguing ‘on their terms’, I can show that. And did. And gave the guy pause. Merely disagreeing over definitions is a waste of time, if the argument is logically invalid. Demonstrating that destroys the arguments but you can only do that if you accept all their terms.

Leave a Reply

Your email address will not be published. Required fields are marked *