Would God's Guidance Be DNA-Detectable?


(John Harshman) #1

So what you’re saying is that the evidence shows us common descent, and faith, not evidence, shows you that it was guided?


More Excuses for Cambrian Non-Evolution
Introducing Bilbo
Does Appearance of Age Render God a Deceiver?
(S. Joshua Swamidass) #2

A better question: if God guided, would we expect to find evidence?

Given the signal to noise ratio, and intractable ignorance of the totality of sequence space and history, the answer should be obvious. We do not expect there to be any evidence.


(John Harshman) #3

I’d agree, with accent on “expect”. Given God, a very ambiguous concept, we can have no idea what to expect or not expect. What we saw might be indistinguishable from purely natural evolution or it could be instantaneous, recent creation of unrelated species with radically different biochemistries. Or every third species could be metallic, living clockwork.


(S. Joshua Swamidass) #4

In this case I’m proposing a specific model of guidance. Common descent plus occasional mutations guided by God that are required to keep evolution on the right track. So this is not an ambiguous concept; it is specified.

What we know is the vast majority of mutations (and variants and divergence) are neutral, out numbering functional mutations by a 1000 fold. We know this by several independent means. If God sprinkled in a few key mutations, we can ask if it would be possible to detect them. The answer is unequivocally “no” unless God had among his design goals to make it clear to us in DNA that he was guiding evolution. However revealing Himself in DNA doesn’t make a lot of sense in theology and is a different goal than guiding evolution. So we can say quite confidently that we do expect to know from DNA that God is guiding evolution.


(John Harshman) #5

Sure, you can propose that, but is there any reason to propose that model other than the fact that it’s compatible with the data and can’t be distinguished from natural evolution? There’s no theological reason to suppose that God would do it that way rather than some other way. One can’t distinguish it scientifically from a deist or atheist view, and in fact that’s its chief reason for being.


(S. Joshua Swamidass) #6

Yes there is reason. People claim that guidance didn’t happen because there is no evidence of it. We can test the logic of such a claim by asking if we would even expect to see evidence of guidance in an obvious model. We do not, and that makes obvious the original claim is just incorrect.

You are a great scientist @John_Harshman, but this is dipping into philosophy and and theology. I do not think this is a valid argument or conclusion.


(John Harshman) #7

That seems like a nice way of telling me to shut up. Nevertheless, I persist. What’s invalid?

OK, you have advanced a notion that’s compatible with the evidence because it was designed to be compatible to the evidence. We can’t reject it because it was designed to be incapable of testing. We can agree that if God acted in this way those actions would not be detectable. We could, however, fall back on the heuristic of Occam’s razor, and God is one serious multiplied entity.


(S. Joshua Swamidass) #8

No, I’m not saying that.

That is false. I did not design it for the purpose of being compatible with the evidence. That is false.

I never claimed it was the simplest explanation. i just suggested it as a hypothesis to test, and asked if the data could discriminate it. The data cannot. That does not rule it out, but rather demonstrates the evidence does not rule it out. From the DNA evidence alone, I do no think we should conclude that God guides evolution either. Rather, that would come from considering additional non-scientific evidence that you don’t see value in. That is fine, you don’t have to agree. However, the DNA evidence does not demonstrate that God never guided evolution. We do not expect to see evidence in DNA of his guidance.


(Ashwin S) #9

The problem with Ockham’s razor is that nothing in life or the universe seems simple or limited to the most parsimonious explanation.


(John Harshman) #10

I would disagree with that. Occam’s razor has proven to be a very successful heuristic in science. There’s a difference between “simple” and “simplest explanation compatible with the data”. The latter can be quite complex, just not more complex than necessary.


(John Harshman) #11

It seems to me that you did. You are limiting God’s actual to small, occasional tweaks that are lost in the noise, and I can’t see any reason for such a limit other than to make the hypothesis compatible with the data. What other reason?

Agreed, just as the data can’t rule out any other hidden variables that have no abserved effect. Maybe what looks like a good fit to a line isn’t really a linear phenomenon but a 5th degree polynomial in which all the non-linear terms are too small to have any noticeable effect. Nevertheless, we generally suppose the the underlying phenomenon is linear.

Which evidence do you refer to?

That’s only true if we posit that guidance is exceedingly subtle. Why should we posit such a thing? Why should we suppose that God wouldn’t have inserted three new chromosomes, pulled from nowhere, into an ape genome in order to create humans? Or chimpanzees, for that matter. I maintain that there is no reason to suppose the subtlety and rarity of intervention other than to comport with the known data that there is no evidence of gross intervention. But I await your reasons.


(Ashwin S) #12

Here the necessity is brought about by the data available. When more data becomes available, the understanding can become more or less complex. For example, take spontaneous generation for example. Totally compatible with Ockham’s razor… but false.
The concept doesn’t have any ontological reality. Things don’t have any need to be as simple as possible.

Edit: There also seems to be some subjectivity involved in deciding what is the “necessary” amount of complexity.


(John Harshman) #13

And yet it’s a good heuristic. If we didn’t use it we could never choose any hypothesis as the best over countless other hypotheses that explain the data equally well. We would be unable to reject “angels push the planets about” as inferior to Newton’s universal gravitation.


(Ashwin S) #14

Except Newton’s universal gravitation proved to be wrong. It was not action from a distance.
I understand it’s value as heuristic. However, that doesn’t make the choice the right one.

Ockham’s razor has nothing to do with whether something is true or not.


#15

That depends entirely on how God guided evolution. God could change every neutral mutation back to what it was and produce convergent mutations throughout the tree of life. I don’t think it is too difficult to come up with scenarios where God could make his actions detectable. God could also make his actions undetectable, but I don’t see why we would expect this when starting from first principles.

When you say that we do not expect there to be any evidence I think that is influenced by what we have already learned about the history of life.


(S. Joshua Swamidass) #16

I’m not limiting God to small occasional tweaks. Rather Im stating what we know from Biology, and the hypothesis of guidance itself. It does not take many changes to have a large impact, so he does not have to make many changes.

If you think that scientific assessment is incorrect, please do make your case.

That’s my point. Science is right to explain biology without God. I’m just saying that from a theological perspective there is no reason to think that we’d find evidence of guidance in DNA. The fact that we do not find evidence, therefore is not evidence that God did not guide or that his guidance was not important. Rather, these are just questions outside the genetic streetlight. We do well to explain the limits of scientific knowledge alongside its findings.

Scripture and theology.

Neutral theory and an immense amount of biological evidence that only small changes in DNA are needed for large changes in phenotype.


(John Harshman) #17

But it does have considerable to do with which scientific hypotheses we prefer. It helps us avoid being unable to choose from infinitely many hypotheses that fit the data.

I wouldn’t say Newton was wrong, either. His main idea was that a simple equation could explain what we see on earth and in space. His equation later needed a little bit of tweaking, but it still works for most purposes. And we still don’t evoke angels to explain planetary motions. Why not?


(John Harshman) #18

But likewise there is no reason to think that we wouldn’t find evidence, right? How do scripture or theology provide evidence of divine intervention in evolution, even for the most charitable interpretation of the word “evidence”? If we grant that god exists, why not a god who set up the initial conditions and let the system run from that point?

I don’t see the relevance of neutral theory. And while it’s true that small changes in genotype can produce fairly large changes in phenotype, it’s also true that large changes in genotype can produce even larger changes in phenotype. Why would an active god necessarily limit himself to those small changes? Why, for example, could he have humans emerge in a single step from an early primate rather than having changes accumulate over tens of millions of years in a long, branched lineage?


(S. Joshua Swamidass) #19

I like some of the fine distinctions you are making. I’ll respond later.


(Jon Garvey) #20

Agree. Noting the crossover is important, because one is making theological conclusions (“We don’t know how God would do it”) from a naturalistic frame of reference (“What does the empirical evidence prove?”) - the same fault that is with justification levelled at ID.

Two points that I’ve made before. The first is on philosophical presumptions: and that is that non-lawlike contingencies may be explained equally well by chance or by choice. The two are formally indistinguishable (something highly improbable might always be the the result of a fluke rather than a decision), but they actually belong to two incompatible metaphysical systems.

You can have Epicurean chance, or Theistic providence, but both together is philosophically problematic. In the first, all non-lawlike events are evidence of chance, but in the second they are all evidence of choice. The matter is not about science, but the world view underlying how you do science.

The second point is on theological presumptions. “We have no idea how God might do things” is based on “God” as a hypothesis about an alien being. But starting from Christianity, we are taught that we were created in the image of Christ, through whom all things were created. As Paul says of believers “We have the mind of Christ,” and that applies to an extent at the level of our humanity, not just our beliefs.

This explains why we can understand so much of the universe, why maths is so unreasonably effective - and why most men see the hand of a creating divinity throughout nature. We are made to correspond to God - indeed the Scriptural narrative is that we were created to govern the world on his behalf. So whilst God’s work may sometimes be counterintuitive, it is not (theologically speaking) something entirely other to us: it is, indeed, the work of Christ, whom we may know personally.