Ashwin on Common Descent


(Ashwin S) #1

Your argument seems to be that common descent explains the data best… and there is no competing alternative.
The question that comes to mind is … what if the data was different (I.e there is no uniform rate of mutation)… would it disprove common descent?. (That’s what a test should mean right… a possibility of failure).

For me, the data seems to say that species diverge and new species stabilise very fast… and we are essentially measuring the genetic drift over time.

Edit: Frankly, if we concede that the only natural way for a living organism to emerge is reproduction from another living organism… Then common descent is the directly connected to an assumption of natural causes.


JoeG's Case Against Common Descent
(S. Joshua Swamidass) #2

Yes that is right.

It is true that common descent is the only natural way for organisms to emerge. But the data also fits what we expect if common descent is true. If it did not, the situation might be very different. It is hard to predict what science would do if there was clear and obvious evidence that humans arose by a non-common descent process. We can imagine what that might be…

  1. Perhaps human genomes could have been totally different from other life, clearly separable.

  2. Perhaps human biochemistry could have been totally discontinuous, as in our amino acids and nucleic acids were the mirror version of all life around us.

These are both examples of data that could have, in principle, definitively proven that humans and apes do not share common ancestors. Science, under its current rules, would like got to more exotic explanations, like aliens and such. As unlikely as that is, they might be forced their if the evidence against common descent were strong. Or maybe then their might be a rule change, specifically for this case. We are talking about a fictional world though, so it is hard to know.

Certainly common descent is falsifiable, at least in principle. It has not been falsified, because the data just matches up so closely. At the very least, we have to grant that it appears we share common ancestors with the great Apes, even if this is not ultimately the case.

Good questions @Ashwin_s


(Ashwin S) #3

Let me try and engage with this. First off, if the calculations are true, then time required for new species to arise from a common ancestor is very limited. This was not the general expectation when common descent was proposed. I think mechanisms of change become crucial here and why change gets triggered and how it is accomplished quickly become crucial. The faster the change, the bigger the teleological challenge.

I would love to understand this more… what level of difference would be satisfactory? And is it fair to only look at the genetic difference as opposed to the gamut of difference in form and behaviour.?
Also, human genetic similarity to animals is not a problem for design. It would be expected because design involves calibrating interactions between systems…and the ecosystem is such a system.

I don’t think this is feasible in a design scenario because did the reason I gave above. The best design would involve biological systems that can interact with and complement each other.

Again, this would be true if the calculations had something to do with the origin/emergence of new species. However , this calculation will work best within a species… It makes sense that the rate of drift in organisms which undergo sexual reproduction would be similar. Or at least correlated.
Isn’t that a consequence of design/process of sexual reproduction? Wouldn’t this rule hold as long as a similar type of reproduction is involved?
Edit: I suspect this mathematical correlation has more to do with mitosis than common descent. It is not causally linked.


(Ashwin S) #4

Thanks… being a non specialist, it took some time to understand exactly what you are getting at…
:slight_smile:


(Ashwin S) #5

@swamidass
Let me propose a thought experiment to explain what I mean.
Let’s assume we have two machines making dolls using largely the same process/steps in different places on earth. Being machines, they will miss the mark and create defects in some cases because of uncontrollable variables.
Since the process followed and the machinery is similar, it’s reasonable to assume defect rates will be similar. We could statistically corelate the defects with a non linear function to account for the variables/minor differences. Higher similarity in the machinery/process will lead to a better correlation… larger differences/independent sources of variations will lead to weaker correlation.

Wouldn’t the above example explain the correlation in mutation rates considering that rats/mice, human beings and chimps all use similar processes to reproduce?
Such a scenario can be explained by design as well as descent.


(S. Joshua Swamidass) #6

Great. Let’s try.

None of these objections seems valid. We’ve observed new species arise in the wild very quickly, in just a single generation. at times. The mechanism of change is largely (but not exclusively) neutral changes. It does not matter what the “general” expectation is. What matters is what the mathematical theory of modern evolutionary science expects.

Differences large enough that it would not be readily explained by known and inferred natural processes. I gave a few examples.

Yes, it is fair. Small genetic changes can produce large differences to form and behavior. That is what is so surprising about the fact that humans and chimps are so much closer genetically than the much more behaivior/form similar mice and rats. That is a puzzle in nature of which neutral theory makes sense.

You are missing the point. Other than common descent, there is not proposed design principle that explains why mice and rates are more different than humans and chimps. If mice and rats can be that different genetically, certainly humans and chimps can be more different. Yet they are not. Why not? The only known design principle that makes sense of this is common descent.

It absolutely is feasible. There is no physical reason this is not possible. All we would need would be some enzymes that would convert between isomeric form. I believe God could have made us that way if He wanted. Why don’t you think He is powerful enough to design us in a way that is obviously impossible to arise from common descent?

They do. So it is true.

We can directly study this process in common descent. Everything seems to line up.

Is there a magical world where species are manufactured that mimics common descent? Sure, maybe. However, no one has produced a valid mathematical theory to explain what we see. This does not, either, change the fact that everything appears to arise by common descent.


(Ashwin S) #7

I will look into this and get back to you at some point. If you can share any links to examples of this, it will definitely help.

This doesn’t make much sense to me. Perhaps you are reading too much significance into genetic drift…

This is not really true. If the species arose quickly, then genetic drift would be a good explanation of the difference. However this would be true irrespective of how the species arose. Even if the first rat was a product of genetic engineering on the CA by aliens, after millions of years, we would get the same data (as long as the time involved is the same)… So how is this significant more than saying life emerges through reproduction. But rather, it seems to take attention from the actual mechanisms of speciation(whatever they are).
I doubt it’s possible to come up with a scientifically acceptable alternative to common descent… Any proposal would involve organisms emerging through supernatural means (either God or aliens concocting life in a test tube).

You missed my meaning. The machine doesn’t represent the process of speciation. It represents the biological machinery/systems involved in sexual reproduction. Given similar systems in rats and humans, there is nothing surprising about getting mutation rates that can be correlated. This would be true of any organisms which propogates similarly via sexual reproduction.
If tomorrow, we managed to create a unique creature
in a lab which propogates via sexual reproduction, we would get the same mutation rate. The only significant factor would be similarity in the machine set up/process used for reproduction. Common descent is a trivial coincidence to this data even if it is true.


(S. Joshua Swamidass) #8

@Ashwin_s, I’m trying to follow you but it is difficult. Can you try reiterating back to me what you think the argument I made is? That might help.


(Ashwin S) #9

OK let me try.
we can calculate the mutation rate for mice/rats.We have estimates for when they diverged as a species. So we can calculate the genetic distance between the two.We can do the same for human beings and the genetic difference difference between humans and chimps. The calculations gives approx 20% for rat/mice and 1.5% for human/chimps. Later on genome studies have validated this finding. And hence things make sense when we look at it from a common descent perspective… and doesn’t make sense in any other perspective because if the mutation rates are constant, then if we extrapolate the data, the mice and rat should converge into a single species just as humans and chimps should.

Is this accurate?


(Ashwin S) #10

This was helpful. If my understanding is correct, I will be able to explain my argument in a more coherent way…
Will post that once you revert back.


(S. Joshua Swamidass) #11

I do not know what this means.


(Ashwin S) #12

If we draw a graph with genetic difference over time… we could extrapolate the current data so that we get an intersection where mice and rats have zero genetic difference… I.e a common ancestor.


(S. Joshua Swamidass) #13

This has nothing to do with the argument I made.


(Ashwin S) #14

Is it a wrong understanding? I don’t think it changes the argument in any way… it just helps in understanding better (at least for me).
Should I try to make my earlier point clearer… will that help?
Edit: For example, if we have the equation D=F1(R,T) for mice and D=F2(R,T) for rats… this can be represented as two curves in a graph which will intersect where we get a CA.


(Ashwin S) #15

@swamidass : On further thought i just want to confirm that we have actual data to make such a prediction. (i.e we are able to map D Vs Time in related species with actual data, fit the trends to an equation and show that the two curves converge more or less in line with fossil evidence/ earlier calculations based on mutation rates. If mutation rates really average out over time then this should be possible)

I got the idea from your thread that this is being done by comparing by “ancient DNA”. Is this a correct understanding?


(Intelligent Design Deist) #16

Mice and rats are not more different than chimps and humans. Rats and mice resemble each other much more than chimps and humans do. The genetic differences between mice and rats are easily explained by evolution by design- designed recombination caused by environmental cues.


(Dr. Patrick Trischitta) #17

LOL not to other rats and mice.


(S. Joshua Swamidass) #18

I want to come back to this. You are getting some of the argument, but you’ve embellished and missed pieces. I don’t think it is intentional, but it may be why we are missing each other. Do you want to try and explain it back to me again? Or should I try condensing it?

I’m not trying to be pedantic, but with a lot of the static from others’ interjections, it is possible the main point was lost in translation.


(Ashwin S) #19

I am not intentionally misunderstanding.
Perhaps you could correct me where I don’t get it as and when you have time…


(S. Joshua Swamidass) #20

I believe you. That is the only reason I’m taking the time to explain it. I’ll take this step by step. Two points first.

First, do you agree that from a “design” point of view apes and humans are much more different than mice and rats?

Second, there is a non-intuitive feature of scientific reasoning that is important here. We care about the explanatory power of a theory more than we care about falsification. I concede up front that no evidence we find contradicts special creation. God could have made things any way He wanted, so no data can falsify special creation. So retorting that God “could have made it that way” is granted but meaningless. We have to show some warranted reason (i.e. a design principle) why God made it the way we see it, and not another way. You have to demonstrate mathematically why special creation follows the D=TR rule for us to think it has equal explanatory power.

Third, quickly, this conversation will begin to interact with both natural theology and special revelation, so that is why we avoid this in science. We can discuss it here, but keep in mind any recourse to theology takes us out of science. That means it might be warranted to believe special creation in theology, but science is still going to teach common descent within the confines of science.

So @Ashwin_s, do you understand these three points? You do not have to agree with #3 or #2, but I want to be sure you understand all the points, and agree with #1.