Also Available in:

Feedback archiveFeedback 2017

Do creationists cherry-pick discordant dates?

A niggling question about radioactive dating

commons.wikimedia.org, Radiogenicthermal-ionization-mass-spectrometer

Radioactive dating is obviously a concern to many because people keep asking about it, even those who have some knowledge of creation science. Today’s feedback comes from David M. of Australia who asks initially about isochrons, but his question is really about radioactive dating in general. What follows is a short conversation between David and CMI geologist Dr Tas Walker, explaining the basic flaw with all dating methods and how geologists manipulate their story after the event:

Hello Tas,

I was wondering if you might be able to help with a niggling question I have about isochrons—that being, “Is it such that the examples presented in creationist materials where rocks of known age, perhaps 50 yrs old, when dated via isochronic methods, and produce results of millions or billions of years of known age—is this normal? Or are creationists guilty of ‘cherry picking’?” [Ed.: Some such relevant articles on isochrons would be Radioisotope dating of rocks in the Grand Canyon and Radiometric dating and old ages in disarray.]

I was reading a non-creationist website where there was an admittance that sometimes results that are ridiculous and discordant are found but that was not the norm as presented by creationists. I, being a teacher and not heavily embroiled in the academic world of the geologist, am not is a position to get insider knowledge. So, discordance and so on—is this normal?

Thanks Tas.

David M.

Hi David,

No, it is not cherry picking. Those sorts of results are normal, but you would not know that by the way the results are presented in the literature.

The key is to understand:

  • It is impossible to determine the age of something in the past by making measurements of properties in the present.
  • Thus, all dates are based on assumptions about the past.
  • The only way a scientist can consider his assumptions are valid is if it gives a date that is acceptable.
  • If it is not acceptable, then in every case he will change his assumptions after he gets his ‘date’ in order to give a plausible explanation for his result. This is called interpreting the result.

This article (The fatal flaw with radioactive dating methods) explains from an everyday example why it is impossible to know the age of something by making measurements only in the present, even if you have a highly accurate timepiece. The point is that you need to know the initial conditions of the parameters you are measuring, and what happened to them all along the way.

This article (The way it really is: little-known facts about radiometric dating) provides a common example of the sort of situation that geologists are confronted with. It shows how the geologists determine the relative ages of rocks from the way they are related in the field. This evidence is quite sound and geologists can be quite confident of this. That is how they know what radioactive ‘date’ would be acceptable. Usually it is a range of dates that would be acceptable (often a large range). This article also explains the sorts of stories geologists use to explain radioactive date away, irrespective of whether it is old or too young. The layman looking on would be none the wiser.

A famous example of the way these after-the-event explanations are invoked and how even the published dates are later changed is illustrated in this article (How dating methods work). It uses a real-life example that played out in the scientific literature over half a decade or more. It recounts the numerous attempts to date a volcanic ash in Africa and how results were rejected because they disagreed with what was considered acceptable. Surprisingly, even after a date was agreed upon it needed to be discarded later after new discoveries came to light. A new date had to be found that could be agreed by the wider scientific community.

It is important to realise that scientists make conflicting dating results ‘agree’ by inventing a story. This is called ‘interpreting the date’. It is a standard part of the dating procedure. Every reputable paper will have a section in which the author will describe how they interpreted their results.

It is not difficult to find articles from the literature that illustrate how, by using this technique, it is impossible to get a ‘wrong’ date. They just interpret whatever date they get so that it sounds eminently plausible. In other words, when they have finished with their interpretation, the dates do not appear to be discordant or ridiculous. Creationists do not play this little game, and so geochronologists tend to regard them as lacking initiative. But creationists examine the raw dates and show that they conflict with what they should be. Evolutionists would not do this but would invent a story such that those dates did not seem discordant at all.

Here is another article that includes examples from the literature showing how dates by different methods that are wildly different are interpreted to seem like they agree: Radioactive dating methods—Ways they make conflicting results tell the same story.

All the best,

Thanks for the reply Tas.

I would love to be privy to conversations held amongst geologists involved in this science, though I wonder if they even know that they are doing what they are doing … massaging the data if you know what I mean.

Yours in Christ


Hi David,

I have been privy to such conversations about dating among geologists, especially during my university days. One of my honours projects was about radioactive dating and we did an exercise about how to interpret dates. As a group of about half a dozen we were presented with a hypothetical geological situation. We were given all the geological details (type of rock, location, size, geological map, etc.) for some half-a-dozen rock samples that were collected. Plus we were given the dating results (xxx.x ± x.x million years) for each sample. All the ‘dates’ were different. Our exercise was to interpret the results—to say what those results meant.

It was very interesting because each participant invented a different scenario as to what the dates meant, and there was a lot of argument about it. The frustrating thing was that no-one could say which scenario was right because no-one was there in the past to know. And the supervisor would not say which interpretation was ‘right’ because there was no ‘right’ interpretation. It was just a matter of who was able to present the most plausible case.

I don’t think anyone involved in that exercise, or those involved in dating professionally, would consider themselves to be “massaging the data”, even though this is what they are doing. They are required to come up with an ‘age’ but they do not have an instrument that can measure it. Rather than say, “I don’t know,” they are simply doing their best to produce something that can be defended.

All the best,

Published: 28 October 2017