Also Available in:
This article is from
Creation 30(3):28–29, June 2008

Browse our latest digital issue Subscribe

How dating methods work


Images from stock.xchng Skull

Addressing the students, I used a measuring cylinder to illustrate how scientific dating works. My picture showed a water tap dripping into the cylinder. It was clearly marked so my audience could see that it held exactly 300 ml of water. The diagram also showed that the water was dripping at a rate of 50 ml per hour.

I asked, ‘How long has the water been dripping into the cylinder?’

Immediately someone called out, “Six hours.”

“Good. How did you work that out?”

“By dividing the amount of water in the cylinder (300 ml) by the rate (50 ml per hour).”

“Excellent,” I said. “See how easy it is to calculate the age of something scientifically? Every dating method that scientists use works exactly the same way. It involves measuring something that is changing with time.”

People began to relax once they understood that the science of dating is not so difficult. Then I surprised them, “The problem is that six hours is the wrong answer.”

They look puzzled and disbelieving.

“I set this experiment up and I can tell you that the water has only been dripping for one hour. Can you tell me what happened?”

Image from stockxpert Skull

After they had composed themselves, someone called out, “The tap was dripping faster in the past?”

“Perhaps,” I said.

“The cylinder was nearly full when you started?”

“Maybe. But can you see what you are doing?” I asked. “In order to calculate an age you made assumptions about the past. You assumed the rate had always been 50 ml per hour and that the cylinder was empty when it started. Based on those assumptions you calculated the time of 6 hours.”

They nodded.

“You were perfectly happy with that answer. Not one of you challenged it.” They agreed.

Scientific dating is not a way of measuring but a way of thinking.

“Then, when I told you the correct answer, do you realize what you did? You quickly changed your assumptions about the past in order to agree with the age I told you.”

Every scientist must first make assumptions about the past before he can calculate an age. If the result seems okay then he will happily accept it. But if it does not agree with other information then he will change his assumptions so that his answer does agree.

It does not matter if the calculated age is too old or too young. There are always many assumptions a scientist can make to get a consistent answer.

Suddenly the lights went on. My audience saw, in a nutshell, the way dating methods work.1 Scientific dating is not a way of measuring but a way of thinking.

How it works in practice

Replica of skull KNM-ER 1470

A layer of volcanic ash in East Africa, called the KBS tuff, became famous through the human fossils found nearby.1

Using the potassium-argon method, Fitch and Miller were the first to measure the age of the tuff. Their result of 212–230 million years did not agree with the age of the fossils (elephant, pig, ape and tools) so they rejected the date. They said the sample was contaminated with excess argon.2

Using new samples of feldspar and pumice they ‘reliably dated’ the tuff at 2.61 million years, which agreed nicely.

Later, this date was confirmed by two other dating methods (paleomagnetism and fission tracks), and was widely accepted.

Then Richard Leakey found a skull (called KNM-ER 1470) below the KBS tuff, a skull that looked far too modern to be 3 million years old.

So Curtis and others redated the KBS tuff using selected pumice and feldspar samples, and obtained an age of 1.82 million years. This new date agreed with the appearance of the new skull.3

Tests by other scientists using paleomagnetism and fission tracks confirmed the lower date.

So by 1980 there was a new, remarkably concordant date for the KBS tuff, and this became the one that was widely accepted.

Which illustrates that, contrary to popular belief, the dating methods are not the primary way that ages are decided. The dating methods do not lead but follow. Their results are always ‘interpreted’ to agree with other factors, such as the evolutionary interpretation of geology and fossils.

References and notes

  1. For more information see Lubenow, M.L., The pigs took it all, Creation 17(3):36–38, 1995; creation.com/pigstook.
  2. Fitch, F.J. and Miller, J.A., Radioisotopic age determinations of Lake Rudolf artifact site, Nature 226(5242):226–228, 1970.
  3. Curtis, G.H., et al., Age of KBS Tuff in Koobi Fora Formation, East Rudolf, Kenya, Nature 258:395–398, 4 December 1975.


  1. For further information see: Sarfati, J., Diamonds: a creationist’s best friend, Creation 28(4):26–27, 2006; creation.com/diamonds, and Walker, T., The way it really is: little-known facts about radiometric dating, Creation 24(4):20–23, 2002; creation.com/dating-reality. Return to text.

Readers’ comments

Bob W.
Could you please provide details of a more accurate scientific method of dating which can tell us the age of the earth more accurately than current scientific methods. For example, if you do not accept dendrochronology as an accurate method or radio carbon dating which method do you accept and and on what grounds. Also, can you identify a non Christian scientist who has provided scientific evidence for a young earth. Thanks.
Tas Walker
Dear Bob,
The only reliable way of knowing the age of anything is by the historical method, by eyewitnesses. Every other method that is based on making measurements in the present amounts to, "What age would you like?" Every result is based on numerous unprovable assumptions.
By the historical method we know that the Treaty of Waitangi was first signed on 6 February 1840. There is no 'scientific' method that could give us that knowledge. By the same method we know the age of the earth from the histroy provided in the Bible, which is accurate and reliable.
What a person thinks about the age of the earth depends on their worldview. So, for you to ask to identify a non-Christian scientist who has provided evidence for a young earth is a contradiction in terms. It's like asking to identify an atheist who argues for the existence of God.
That said, there is much evicence that non-Christian scientists report that contradict the idea of long ages, althought they would not see it that way. See for example 101 evidences for a young of the earth and the universe.
John C.
"Scientific dating is not a way of measuring but a way of thinking."
Science is a way of thinking and of course does require assumptions. It seems to me that while creation ministries likes to question current scientific assumptions, they put forward no answers that are not subject to the same, if not more, criticism.
Tas Walker
Hello John,
It's not "science" but "evolutionary science" that we are talking about here. And while we call the assumptions behind the long-age evolutionary creation myth into question, we also provide an alternative. We point to biblical history as reliable and accurate, and we do our science within that framework. Of course this involves assumptions, but at least the assumptions are out in the open and we can have a productive discussion on the real issues involved—our presuppositions.
Robert M.
I always find it interesting how difficult it is for those of an evolutionary standpoint to even question the dogma they are taught. The evolutionary 'Scientific method' seems to screen out any possible variant facts as irrelevant. With reference to the theory of evolution, "if in doubt, throw it out".
Gary F.
In many universities you are taught "what to think" and not "how to think". It amazes me that I still hear all the regular mantras I heard in college as an unbeliever, "all real scientists believe in [macro] evolution." Groupthink is alive and well.
David V.
This water dropping into a cylinder - explanation to refute dating was quite "simple" indeed. Of course, the further you go back in history the greater your statistical errors will be... but combining several dating methods with fossils and evo-devo approach should give one an acceptably accurate idea of when something existed - at least the order of magnitude would be trustworthy (look at the example they gave underneath... it said... one group found 1,82 mill. .. beforehand there were others who found 2.61 and 3 mill..... so that's not to bad.... at least one would then assume with some confidence it should have occured between 1 and 4 million years ago... that's something ! You'd know it wasn't 15 mill. y.ago.. or even 1 billion y . ago...) (if that's the Worst example a creation-site can find to discredit these kind of measurements... then I must say these methods are pretty much awesome ;D !!!! )

Tas Walker
Hi David,
It seems that you have missed the point. All dating methods are based on assumptions about age, not measurements of age. And you can make the result come out to anything you like depending on the assumptions that you make. Further, comparing results based on one set of assumptions with results based on another set of assumptions does not make the results objective. It just says that the assumptions are using the same beliefs. And in your little 'analysis' you seem to have missed the first 'dates' that came in at 212–230 million years. That changes your story about order of magnitude being trustworthy.
Joshua F.
I agree - this is a very good way to teach students how radiometric dating works. However, there is a caveat that should be mentioned: each of the assumptions scientists use to determine the rate can be tested. We can determine whether the rate was faster in the past, based on other elements that are present near the sample to be dated. In science, it extremely important to challenge assumptions; it is also important to be willing to change those assumptions when challenging data suggests we do so. In this way science is self correcting.

Notice the date of the articles - the "incorrect" dates were published in 1970. 5 years later, the dates were revisited and tested again, this time by other scientists. Also notice that none of the dates support the idea that it all happened 4,500 or so years ago.

I am curious, in your "how it works in practice" example, how you would arrive at a date of 4,500 years? What method would you use? If you were to then submit your methods of analysis to peer review, don't you think the reviewers would want to know why your method differs so much with what is now widely accepted and why your method is better than the one in current use?
Tas Walker
Hello Joshua,

Actually, those assumptions about the past cannot be tested because the past is not available for scientific observation. Scientists can find other evidence in the present that supports or contradicts the assumptions, but such evidence can always be 'harmonised' by developing new hypotheses.

The dates do not support a date of 4,500 years because those doing the tests were not looking to support that date.

The date of 4,500 years has been obtained by the only method that can yield reliable, precise dates—the historical method. That is how we know the date for the Battle of Trafalgar and the Battle of Hastings, for example. That is how you know your age.
Nate L.
Brilliant idea for use in a classroom setting to teach this very idea. I'll be using this in the future for our students! Thanks!

Comments are automatically closed 14 days after publication.