Historical evidence trumps extrapolation of observational evidence

One thing I have appreciated about Swamidass’ book is his commitment to communicating the uncertainty of scientific conclusions. This is something we do not often see from popularizers of science. He repeated this message forcefully at the debate with Behe, saying that science doesn’t have answers to most of the questions the audience is asking. This is consistent with his criticisms of Behe, if not his credulity toward evolutionary explanations. It is quite easy to say science has no answers in criticism of ideas you disagree with. The trick is to form a consistent philosophy of science that allows science to operate in pursuit of new knowledge while acknowledging uncertainty. But considering that most do not publicly acknowledge uncertainty at all I would say Swamidass is making an important contribution to the public debate here.

I noted in particular a passage towards the end of Chapter Two of his book where he is talking about the resurrection of Jesus:

“As an example, consider the Resurrection for a moment. Yes, in every observable example, people dead in the grave for three days do not rise again…There certainly is no scientific evidence against the Resurrection. Science does not do well with singular, localized events in the distant past. In the case of the Resurrection, however, an immense amount of historical evidence points in its direction.”

Swamidass, S. Joshua. The Genealogical Adam and Eve (p. 29). InterVarsity Press. Kindle Edition.

Swamidass is affirming a distinction YECs have made for a long time: there is a difference between observational science and extrapolating observational science into the distant past, which we call historical science. Kudos to Swamidass for acknowledging the distinction.

1 Like

Past events can leave evidence in the present. YEC, for instance, is falsified by ubiquitous amounts of observational physical evidence for an old age of the Earth. On the other hand, a singular, momentary localised event like the Resurrection would hardly be expected to leave behind any observable physical evidence at all.

ETA: whether the historical evidence for the Resurrection is more than just evidence for a group of people holding a particular belief remains an open question.



If this distinction was more consistently honored back when Darwin was alive, the YEC movement would be much smaller today.

Back then, Scientists probably thought “everybody” would move to a figurative understanding of the Bible; the Adventists surprised everyone!

YECs don’t deny that past events can leave evidence in the present. What we deny is that extrapolations of contemporary observations can always be extrapolated into the deep past.

So for example the scientific argument vis a vis the Resurrection goes like this:

A) All humans observed to die today remain dead

B) Therefore all humans in the past who died remained dead

C) Jesus was a human and was observed to die in the past

D) Therefore Jesus did not rise from the dead.

Swamidass and all Christians must deny D, and we universally do so by denying B which is the extrapolation of current observational data into the past. We certainly don’t deny A or C. Swamidass, again following most Christians, denies B by citing historical evidence.

This is exactly what YECs are doing. So for example:

A) Radioactive decay is observed to occur along an exponential curve relating to time.

B) Therefore radioactive decay has always occurred along this same exponential curve in relation to time.

C) Therefore some rocks showing advanced stages of exponential decay must be millions or billions of years old.

YECs deny B, which is the extrapolation of currently observed data into the past, and to do so we cite historical evidence. It is exactly the same reasoning.

This does make the prior probability of the resurrection astronomically low. I mean god could make me turn into a carrot right now. But it doesn’t mean that it’s probable he will. God tends not to turn people into carrots.


That’s not the argument being made. There are a ton of premises you are leaving out. Independent methods coming to the same date. Same species being found in different formations coming to the same date, etc. There are other independent reasons for believing B besides A.


@swamidass there isn’t though. You have a set of facts that aren’t entailed by the resurrection. So these facts are no where near enough to overcome the astronomically low prior probability of the resurrection


Currently reading Swamidass’ origin story which includes the following statement:

pg. 2 of the document, pg 166 of the journal (?) it’s in. Bottom of the page.

No, this is quite false. There are very good reasons to believe that radioactive active decay in the past has been the same as it is now. These are not inductive reasons, but deductive ones based on the consilience of numerous lines of evidence.


So are you saying there’s only one line of evidence that people don’t rise from the dead and therefore this can be denied in a different way than radiometric decay extrapolations?

It’s not quite like that. Nobody acknowledges that extrapolations of contemporary observations can always be extrapolated into the deep past. Mainstream scientists acknowledge that some rates of change can and will fluctuate, and they perform exhaustive studies to determine the exact limits to which different rates can change with time.

The mistake made by YECs is that their ideas of which rates can be extrapolated and which rates can not are completely out of touch with reality. They propose, for example, that nuclear decay rates could have been a billion times higher in the recent past, despite their own admissions that if such changes had any basis in reality, it would have raised the temperature of the Earth to 22,000°C. In reality, scientific studies (astronomical observations as well as laboratory-based experiments) have placed very tight constraints on the extent to which nuclear decay rates can vary in reality, and those constraints are just a fraction of a precent at most. At the same time, rates of change that are very poorly quantified and subject to very wide variation – such as, for example, the Earth’s population, changes in the Earth’s magnetic field, or the rate of influx of salt into the ocean, are held up as being more reliable than the high-precision methods. I’m sorry, but measurement simply doesn’t work that way.

That doesn’t make sense. People rising from the dead have no bearing whatsoever on nuclear decay rates. The two are completely separate issues.

I believe in the Resurrection. But billion fold accelerated nuclear decay with the 22,000°C worth of heat that it generates being miraculously removed for no reason whatsoever other than to make the earth look older than it really is in the most complicated and convoluted way imaginable? Give me a break.


None of this contests my point that both arguments are exactly the same logically. You are arguing that the weight of evidence is different for the two cases. Fine. But the form of the argument is exactly the same.

I agree with what you say here, but actually the situation is even worse. Under the assumption of constant decay rates we have measured ages that vary in the range of some thousands of years (C14) to thousands of million of years (U-Pb, or Rb-Sr for instance). If YEC is correct, all the stuff we measured should have pretty much the same age: 6000 years, give or take.

To arrive at our measured dates the rates should not only have been very much larger than they are now, they should also have varied in significant amounts (as in, orders of magnitude) from sample to sample, so that we now see the spread in dates that we see. All the while preserving the alignment between different aliquots of the same sample. Furthermore, where we have analyses of the same sample using different decay chains and therefore different rates, the variations of these specific rates would have to be exactly fine-tuned so that they still end up yielding the same dates!

You would be better off calling it all a miracle and be done with it.


It doesn’t work that way. The weight of the evidence is more important than the form of the evidence. Large samples with small error bars take precedence over small samples with large error bars. Period. End of story.

That is simply how measurement works in every area of science.


Maybe it is more important. But the point is that sometimes historical evidence outweighs scientific extrapolations into the distant past and this is widely recognized and used by serious people.

The YEC denial of premise ‘B’ in case of radiometric dating flies in the face of all evidence. On the other hand, Resurrection deniers conclude Resurrection premise ‘B’ with support of all evidence. So no, the arguments are not the same at all.

BTW personally I wouldn’t follow your reasoning to deny the Resurrection because I don’t think it is valid, but that is a separate story.


Which is exactly what the Resurrection is: a miracle.

We are using the same evidence in both cases: historical texts claiming to recount what happened.

If that is what you believe, fine with me. Then why not claim that the appearance of old age is a miracle too?

The Bible doesn’t record a miracle here as it does with the Resurrection so it’s better to try and find a scientific solution. In fact, it may even be a clue to greater scientific understanding of nature than we currently have, and working to solve problems like this might lead to scientific discovery in the future.

Which by the way, is a possibility even if its incorrect, which is why YECs shouldn’t be ruthlessly excluded from science.