How archaeologists date prehistoric evidence
A headline may say that a tool is thousands of years old, but the laboratory may have measured charcoal beside it rather than the tool itself. That distinction does not automatically invalidate the result. It tells us which relationship must be explained before the age can support a historical claim.
This guide follows a simple chain: sample, measurement, association, interpretation. The aim is to help you read an archaeological report critically, not to identify or excavate artifacts yourself.
1. Start with context
A find’s position can tell a different story from its appearance. The Smithsonian’s dating overview distinguishes relative ordering from techniques that estimate numerical ages. Stratigraphy studies the sequence of layers, but movement and disturbance complicate a simple “deeper means older” rule.
Consider a fictional trench. Layer A is above layer B. If both layers are intact, the sequence can support an ordering of deposition. If a later pit cuts into B, an object in the pit may be younger than the material surrounding it. A photograph of the object alone cannot resolve that relationship.
2. Ask exactly what was dated
Radiocarbon estimates concern carbon-bearing samples. The Oxford Radiocarbon Accelerator Unit describes an approximate usable span reaching back around 50,000 years, depending on the sample and measurement. It is not a technique for directly dating a stone handaxe or a dinosaur fossil tens of millions of years old. See Oxford’s laboratory methods overview.
In the fictional trench, imagine charcoal below a tool and charcoal above it. Those samples may help constrain a sequence only if their relationship to the deposit is sound. Old wood, later contamination, or redeposited material can change the interpretation. The central question is not simply “Was there a lab result?” but “Why does this sample date the event being discussed?”
3. Do not confuse radiocarbon years with calendar years
Radiocarbon measurements conventionally use BP, with “present” fixed at 1950. Atmospheric carbon variation means a measured radiocarbon age requires calibration before it is expressed as a calendar-age estimate. A result can correspond to more than one calendar interval. Oxford explains this in its calibration guide and example laboratory-result explanation.
When a news story reduces a calibrated range to one rounded number, information has been compressed. That may be acceptable for an introduction, but the original range and stated probability should remain accessible. “About 10,000 years old” should not become a precise anniversary.
4. Match the method to the material
Some very old sites are constrained using dated geological material rather than carbon from the artifact. Other techniques address different physical processes. The Smithsonian’s overview includes several methods and their limits. A method name is not a universal stamp of certainty; the relevant material, assumptions, and relationship to the find still matter.
For reading purposes, keep two labels beside every age: measured material and event inferred. A result can be technically well measured while its association with the event remains debatable. Conversely, several independent lines of evidence may support an interpretation better than one apparently precise date.
A fictional evidence worksheet
| Claim | Question to ask |
|---|---|
| “The stone tool was radiocarbon dated.” | Was the sample actually stone, or associated organic material? |
| “The deepest object is the oldest.” | Is the layer intact, and is there evidence of a later cut or movement? |
| “The date is exactly 8,000 years.” | What are the units, calibration, range, and uncertainty? |
| “One discovery proves a migration.” | What links the find to a population movement rather than another explanation? |
5. Practice without disturbing a site
Choose a museum’s online object record or an open research summary. Write down the object, its context, the dated sample, the method, and the authors’ stated limitation. Mark any field that is missing rather than filling it with a guess.
Then write two sentences: one reporting the observation and one describing the interpretation. For example, “The report describes charcoal from a recorded layer” is an observation about the report. “The charcoal supports the proposed occupation date” is an inference that depends on context. Keeping them separate improves both history writing and scientific reasoning.
Where to go next
Return to Prehistory in perspective for the broader timescale. Follow the institutional references above for the methods behind specific claims. This guide contains no new dates, excavations, or laboratory measurements.