Radiometric Dating Methods: Q&A

Radiometric Dating Methods: Q&A

Q:
Is it correct to say that all radiometric decay‑rate dating methods rely on assumptions about the original state of the sample and external conditions that could affect decay, and that these assumptions may introduce errors large enough to skew results by millions of years?

A:
Radiometric dating is built on three foundational assumptions. Each one is necessary for the method to function. Each one is also unknowable in absolute terms. When any of these assumptions fail, the resulting age can be, is, dramatically wrong.

Assumption 1: The Initial Conditions Are Known.

Radiometric dating begins by assuming the starting ratio of parent to daughter isotopes.

But no scientist was present when a rock formed.
No one observed the initial isotopic composition.
No one can reconstruct it with certainty.

Geologists often infer the starting ratio based on mineral behavior, for example, assuming zircon crystals exclude lead when they form. But this is still an assumption. If the starting ratio is wrong, the calculated age is wrong.

There is no way to know the initial isotopic state of a rock that is allegedly billions of years old. Even thousands of years is beyond direct observation.

Assumption 2: The Sample Has Remained a Closed System

Radiometric dating assumes that nothing has entered or left the sample since its formation.

But rocks do not sit in laboratory isolation.
They are heated, cooled, fractured, weathered, buried, exposed, and moved.
They experience pressure, vacuum, groundwater, chemical alteration, and tectonic stress.

Each of these can add or remove isotopes.

We cannot know the full environmental history of any rock.
We cannot know whether isotopes migrated, escaped, or were introduced.
Yet the method requires that we assume none of this happened.

If the system was not closed, the date is not reliable.

Assumption 3: The Decay Rate Has Always Been Constant.

Radiometric dating assumes that decay rates have never changed.

This is treated as certain, but it is still an assumption. Decay rates are measured in the present and projected backward across millions or billions of years.

If decay rates varied even slightly, the calculated ages would shift dramatically.

We cannot directly measure ancient decay rates. We can only assume they were identical to today’s.

Potential Sources of Error.

Because the method rests on assumptions, several failure points exist:

Incorrect initial conditions: wrong starting ratios, wrong age.
Contamination or alteration: open system, wrong age.
Environmental effects: pressure, heat, fluids, or chemical changes altering isotopes, wrong age.
Decay‑rate variability: even minor changes produce massive shifts in calculated time, wrong age.

These are not fringe concerns. They are acknowledged within the method itself.

So, does this sound like good science?

Radiometric dating is often presented as a precise, objective measurement , a clock that simply ticks forward. But the method is not a clock. It is a calculation built on:

assumed initial conditions
assumed closed systems
assumed constant decay rates

Assumed, assumed, assumed.

When the foundation is assumption, the conclusion cannot rise higher than the foundation.

Radiometric dating measures the present.
It infers the past.
And inference is only as strong as the assumptions beneath it.

~Tony

© A.K. Pritchard 2020, 2026 – Free to use with proper attribution.

Anthony Pritchard said:

Radiometric Dating Methods: Q&A

[snip] ... does this sound like good science?

Radiometric dating is often presented as a precise, objective measurement , a clock that simply ticks forward. But the method is not a clock. It is a calculation built on:

assumed initial conditions
assumed closed systems
assumed constant decay rates

Assumed, assumed, assumed.

When the foundation is assumption, the conclusion cannot rise higher than the foundation.

Radiometric dating measures the present.
It infers the past.
And inference is only as strong as the assumptions beneath it.

~Tony

© A.K. Pritchard 2020, 2026 – Free to use with proper attribution.

Click to expand...

I’m no nuclear physicist but I’ve worked in a medical nuclear lab for more than two decades prior to retirement.

First, there isn't just one thing called “radiometric dating.”
There are numerous independent dating systems with different assumptions and failure modes.

Radioactive dating is an established science which relies upon the measured decay rate of an isotope.
When the appropriate methods are used on a suitable sample, the method can be extremely reliable.

Radioactive dating is not infallible. There are weaknesses that may effect the results of the testing.
You’ve listed a few of the “assumptions” and inferred that these are fatal flaws.


It is true that radioactive dating has analytical uncertainties and is vulnerable to a variety of geological factors.
But these vulnerabilities are not hidden.
Recognizing and detecting the processes that introduce error is a part of the science of geochronology.

Researchers don’t generally rely upon one isolated measurement to determine the age of a sample or site.
And different radioactive systems can also be used to cross-checked geological material.

With more recent samples, additional methods may be used; for example, dendrochronology (tree-rings).

Yet, it is important to note that when a date for a sample is determined there is always a ring of uncertainty around it - it's a best estimate answer - the further back a date is estimated, the larger the ring of uncertainty.

A most reliable estimate is determined by using:

1. a variety of independent researchers
2. a variety of independent methods using different physical principles
3. appropriate materials
4. agreement within their uncertainties
5. in a geological context that makes sense with previous estimates.

And even then, occasionally, out of the blue, something arises that disturbs every known theory.

That's how science works...

Rob

Deacon said:

I’m no nuclear physicist but I’ve worked in a medical nuclear lab for more than two decades prior to retirement.

First, there isn't just one thing called “radiometric dating.”
There are numerous independent dating systems with different assumptions and failure modes.

Radioactive dating is an established science which relies upon the measured decay rate of an isotope.
When the appropriate methods are used on a suitable sample, the method can be extremely reliable.

Radioactive dating is not infallible. There are weaknesses that may effect the results of the testing.
You’ve listed a few of the “assumptions” and inferred that these are fatal flaws.


It is true that radioactive dating has analytical uncertainties and is vulnerable to a variety of geological factors.
But these vulnerabilities are not hidden.
Recognizing and detecting the processes that introduce error is a part of the science of geochronology.

Researchers don’t generally rely upon one isolated measurement to determine the age of a sample or site.
And different radioactive systems can also be used to cross-checked geological material.

With more recent samples, additional methods may be used; for example, dendrochronology (tree-rings).

Yet, it is important to note that when a date for a sample is determined there is always a ring of uncertainty around it - it's a best estimate answer - the further back a date is estimated, the larger the ring of uncertainty.

A most reliable estimate is determined by using:

1. a variety of independent researchers
2. a variety of independent methods using different physical principles
3. appropriate materials
4. agreement within their uncertainties
5. in a geological context that makes sense with previous estimates.

And even then, occasionally, out of the blue, something arises that disturbs every known theory.

That's how science works...

Rob

Click to expand...

Let me clarify the scope of my point, because I am not addressing laboratory practice or the professionalism of researchers. I am addressing the logical structure of the method itself.

Radiometric dating does not directly measure age. It measures present isotope ratios and then infers a past based on three necessary assumptions:

• the initial conditions are known
• the system remained closed
• the decay rate has been constant

Those assumptions may be standard, but they remain assumptions, not observations. The reliability of any calculated age is therefore limited by the certainty of those assumptions.

Cross‑checking, multiple methods, and geological context do not remove the inferential nature of the process.

My point is that it radiometric dating is not a clock. It is a model built on unobserved starting conditions and untestable historical continuity.

That is the distinction I am drawing.

Anthony Pritchard said:

And inference is only as strong as the assumptions beneath it.

Click to expand...

Radioactive dating is a model-based inference to measure elapsed time from measurable isotope ratios.
Its reliability depends on assumptions concerning initial conditions, system closure, decay behavior, and the geological history of the sample.

The resulting age estimate is not a direct observation of elapsed time, but an inferred time.

On this we agree.

Radiometic dating deals with the philosophical questions regarding how to measure the unobserved, unrepeatable past.
It does this by observing of the consequences of such an event.

Properly chosen methods can reduce, estimate, or independently test many of these assumptions
The strength of its conclusion depends on how well those assumptions are justified, tested, and corroborated by independent evidence.

Additionally, having an awareness of what the methods actually measure helps an investigator to form the right question thus clarifying the answer.
The model doesn't always answer the simple question desired; (i.e., How old is this rock?)

"You must ask the right questions."

Radiometic dating provides a time-period for the samples...

• crystallization,
• metamorphism,
• cooling below a samples closure temperature,
• mineral growth,
• alteration,
• resetting of an isotope system,
• eruption,

“That, Detective, is the right question.” (Dr. Alfred Lanning in I, Robot)

Asking the wrong question produces a worthless answer.

A good example of asking the wrong question can be observed by reading Young Earth Creationist sites pointing to dating errors of Mount St. Helen.

Rob

Deacon said:

Properly chosen methods can reduce, estimate, or independently test many of these assumptions

Click to expand...

I appreciate the agreement on the inferential nature of radiometric dating. But the central issue remains: the assumptions are not merely difficult, they are unknowable in principle. Managing assumptions is not the same as justifying them.

Deacon said:

Radiometic dating deals with the philosophical questions regarding how to measure the unobserved, unrepeatable past.

Click to expand...

Calling this a “philosophical question” softens what is actually a hard limitation of empirical science. Science is constrained by three criteria: it must deal with what is observable, testable, and repeatable. Radiometric dating requires assumptions about initial isotope ratios, system closure, decay constancy, and geological history. None of these can be observed, tested, or repeated. They are not facts; they are unverified premises.

Methodological refinements cannot justify these assumptions. The past event itself cannot be observed, cannot be repeated, and cannot be independently verified. That is not philosophy. That is the boundary of what science can know. When assumptions are treated as established facts, the conclusion inherits their uncertainty.

In simplistic terms when I first heard about this method of dating the earth, the comparison say of lead to uranium using the decay rate, I asked what if there was some lead in there to start with. I was told that there wouldn't have been.

Seemed like a pretty big assumption to me.

When modeling U - Pb radioisotope dating, the assumption is that there is no initial daughter element.

In real-life U - Pb dating researchers apply techniques specifically designed to deal with initial Pb.
One method, concordant dating, isolates the various Uranium isotopes, their measured ratio can help to determine how open or closed a sample might have been.

Interesting side note:
When working around radiation sources workers sometimes use lead aprons and vests to protect themselves from harmful radiation sources.
Back sometime in the mid 1990's our lab received a shipment of gear from Chile that contained a radioactive substance.

Over the time I worked in the lab we used a variety of manufactured isotopes, first thallium, then technetium, and now, rubidium, each having shorter and shorter half-lives.

Thallium has a half-life of about 3 days. It takes about 10 half-lives to clear the system.
Once a young man came in for a stress test. the next day his father took him to Washington D.C. and they joined a White House tour.
While waiting in line, the kid said that suddenly doors opened and he and his father were pulled out of line and questioned. --- We got a call from the WH team that day!

Rob

Deacon said:

When modeling U - Pb radioisotope dating, the assumption is that there is no initial daughter element.

In real-life U - Pb dating researchers apply techniques specifically designed to deal with initial Pb.
One method, concordant dating, isolates the various Uranium isotopes, their measured ratio can help to determine how open or closed a sample might have been.

Interesting side note:
When working around radiation sources workers sometimes use lead aprons and vests to protect themselves from harmful radiation sources.
Back sometime in the mid 1990's our lab received a shipment of gear from Chile that contained a radioactive substance.

Over the time I worked in the lab we used a variety of manufactured isotopes, first thallium, then technetium, and now, rubidium, each having shorter and shorter half-lives.

Thallium has a half-life of about 3 days. It takes about 10 half-lives to clear the system.
Once a young man came in for a stress test. the next day his father took him to Washington D.C. and they joined a White House tour.
While waiting in line, the kid said that suddenly doors opened and he and his father were pulled out of line and questioned. --- We got a call from the WH team that day!

Rob

Click to expand...

Rob, thank you for the detailed description of U‑Pb procedures and the practical challenges of working with isotopes. The techniques you mention, including concordia diagrams and attempts to account for initial lead, are well known. My concern is not with the skill or care of modern technicians, but with the unavoidable assumptions that underlie every radiometric dating method.

Even when researchers attempt to correct for initial daughter isotopes, they must still assume that the correction model itself reflects the true initial conditions. They must also assume that the sample has remained closed, that no parent or daughter isotopes have entered or left the system, and that the decay rate has remained constant for the entire history of the sample. These assumptions cannot be verified for ancient materials. They can only be inferred, and the inferences depend on the very models being tested.

Your anecdote about the young man triggering radiation detectors is interesting, and it illustrates how sensitive modern instruments are. It also shows that even in controlled environments, unexpected variables can appear. In radiometric dating, the variables are not merely unexpected but entirely unknown, because the initial conditions and the full history of the sample are inaccessible.

My point is simple. We can measure the present ratio of parent to daughter isotopes with great precision. What we cannot know is the original ratio, the full provenance of the sample, or whether the decay rate has been constant across the entire timespan being assumed. The methods are precise, but the precision rests on assumptions that cannot be independently confirmed.

That is the heart of the issue.

A brief clarification on the purpose and implications of the article:

The purpose of the article was to show that every radiometric dating method rests on assumptions that cannot be verified for ancient materials. The methods measure present isotope ratios with great precision, but they cannot determine the original ratio, the full history of the sample, or whether the decay rate has remained constant for the entire timespan being inferred. Because these assumptions cannot be independently confirmed, the calculated ages cannot be treated as reliable indicators of true age.

This is why radiometric dating cannot establish deep time. The appearance of precision in the numbers does not remove the uncertainty in the assumptions. The methods are mathematically consistent, but the consistency rests on premises that are unknowable. That is the central point of the article, and it applies to all radiometric dating techniques, regardless of the specific isotopes or correction models used.

True Parity Between Evolution and Creationism

There is true parity between evolution and creationism at the level of faith. Both systems begin with assumptions about origins that cannot be proven by direct observation. Creationism begins with the authority of Scripture. Evolution begins with the authority of naturalistic premises such as deep time, uniform processes, constant decay rates, and closed systems in the distant past. Neither position can appeal to repeatable observation of origins, because the events in question are not accessible to experiment.

The parity between evolution and creationism concerns epistemological faith in unobservable past events, but biblical faith has substance and evidence because it rests on God’s revelation, not on human assumptions.

This is why radiometric dating cannot settle the question. The present data can be measured, but the past conditions must be assumed. When the assumptions are untestable, the conclusions rest on faith in the framework that interprets the data. Deep time is not measured; it is inferred through a belief system that assumes what the past must have been. Creationism and evolution both operate from faith commitments, but they place their faith in different authorities.

Anthony Pritchard said:

Your anecdote about the young man triggering radiation detectors is interesting, and it illustrates how sensitive modern instruments are. It also shows that even in controlled environments, unexpected variables can appear. In radiometric dating, the variables are not merely unexpected but entirely unknown, because the initial conditions and the full history of the sample are inaccessible.

Click to expand...

Concerning the sensitivity of modern equipment...
Technetium has a much shorter half-life. The effective half-life is around 24 hours.
Unfortunately we have a large trash dump site in a nearby town with very sensitive radiation detectors.
The trash truck drivers routinely set off the alarms, as they drive past the detectors in their trucks ... so they ended up loosing 2 days work for their stress test.

~~~~~~~~~

Unknown and inaccessible conditions do not mean the estimates are worthless.
There are other ways to date geologic sites.
Collaboration with other dating methods may support collected data.

The date estimates are not hard and fast; there may be some wide variability among labs testing facilities - and occasionally there are even new methods that refine or replace older technology.
This is understood by those using the data.

It's a bit like solving a mathematical equation with several unknown variables — the mathematical calculation can be exact while the variables (or assumptions) remain uncertain.

...or like using a wrist watch... you make many simple assumptions when you use it, but it gives you a general estimate of the time, enough for you to base your schedule on.

I guess the question for you would be:
How would you date an unknown geologic sample?

Rob

Deacon said:

Concerning the sensitivity of modern equipment...
Technetium has a much shorter half-life. The effective half-life is around 24 hours.
Unfortunately we have a large trash dump site in a nearby town with very sensitive radiation detectors.
The trash truck drivers routinely set off the alarms, as they drive past the detectors in their trucks ... so they ended up loosing 2 days work for their stress test.

~~~~~~~~~

Unknown and inaccessible conditions do not mean the estimates are worthless.
There are other ways to date geologic sites.
Collaboration with other dating methods may support collected data.

The date estimates are not hard and fast; there may be some wide variability among labs testing facilities - and occasionally there are even new methods that refine or replace older technology.
This is understood by those using the data.

It's a bit like solving a mathematical equation with several unknown variables — the mathematical calculation can be exact while the variables (or assumptions) remain uncertain.

...or like using a wrist watch... you make many simple assumptions when you use it, but it gives you a general estimate of the time, enough for you to base your schedule on.

I guess the question for you would be:
How would you date an unknown geologic sample?

Rob

Click to expand...

Thank you for the additional anecdotes. They illustrate how sensitive modern instruments are, but they do not address the core issue. My argument is not about laboratory technique but about the assumptions required to project present measurements into an unobservable past. Cross‑checking methods that share the same assumptions does not remove the assumptions. Unknown initial conditions, unknown system history, and unverified decay constancy do not make the estimates worthless, but they do make them unverifiable. A mathematical model can be internally consistent while still resting on premises that cannot be confirmed. That is why radiometric dating cannot establish true age.

As for dating an unknown geologic sample, I would not treat a number derived from untestable assumptions as a measurement of actual age. Present data can be measured; past conditions must be assumed. The conclusion cannot rise higher than the assumptions beneath it.

Anthony Pritchard said:

The conclusion cannot rise higher than the assumptions beneath it.

Click to expand...

I think I've communicated that the results of radiometric dating are not a strong, certain declaration of time but rather point towards an ancient creation... the time estimates are quite broad, fitting the known data.

You have listed assumptions and stretched them beyond reason in order to support "Creationist" assumptions.
The scientific data does not support the premise: You are a Creationist by faith.

But God has not asked us to believe in Creationism; he's asked us to believe in the Creator.

Rob

Deacon said:

oint towards an ancient creation... the time estimates are quite broad, fitting the known data.

Click to expand...

God's Word is to be believed. John 17 verse 17, “Sanctify them through thy truth, thy word is truth.”

The Scriptural pattern does not point to deep time. It presents instantaneous maturity from the moment God speaks. I have addressed this fully here:

The Lie of Deep Time The Lie of Deep Time