In effect, this means that some of the fish swimming in the Trave today seem to be over 2000 years old, when radiocarbon dated.
In effect, this means that some of the fish swimming in the Trave today seem to be over 2000 years old, when radiocarbon dated.

This is another example of faulty assumptions and problems with carbon-14 dating. – Ed.

Bente Philippsen & Rasmus Rørbæk / 19 DEC 2014 (25 MAR 2013) – Some years ago in northern Germany, archaeologist Sönke Hartz carried out excavations at a prehistoric camp-site belonging to the Ertebølle culture, close by the river Trave.

During these excavations he discovered an ancient pottery sherd which held remnants of burnt food. Hartz, an expert in the Stone Age of northern Europe, sent the pot sherd away for carbon-14 dating and was amazed when the laboratory came back with a date of 5200 BC.

“It was an archaeological sensation! This pottery was many hundreds of years older than all the pottery that was previously found in Northern Germany. It was older than everyone expected. But, then I was in doubt. I had found the pot by the river, so the food crust could possibly consist of fish. I remembered that there were dating problems with freshwater fish, which could give misleading ages,“ explained Hartz.

Reservoir effect

In order to obtain a radiocarbon date, the amount of remaining Carbon-14 atoms in a sample are measured. The less Carbon-14 that is left, the older the sample.

Hard water contains less Carbon-14 than the atmosphere, because dissolved carbonates are Carbon-14 free. A fish caught in hard water has thus a higher Carbon-14 age than contemporaneous terrestrial samples. If such a fish is then cooked in a ceramic pot, the radiocarbon age of the food crust will be higher than if a terrestrial animal was cooked in the pot.

This is known as the “reservoir effect” because the fish’s carbon actually comes from another “reservoir” than the carbon in terrestrial animals from the surrounding area. “Reservoir age” is the difference between the true age and the Carbon-14 date.

The effect, highlighted by the erroneous date from the carbonised residue on Sönkes’ ceramic sherd, persuaded The AMS 14C Dating Centre at Aarhus University in Denmark that they needed to carry out further investigations.

Variety and size of error surprising

On examining freshly caught fish from the River Trave the results revealed not only a large reservoir effect, but also a dramatic variance from between 500 to 2100 years. In effect, this means that some of the fish swimming in the Trave today seem to be over 2000 years old, when radiocarbon dated.

Felix Riede, an archaeologist at Aarhus University who regularly uses Carbon-14 dating in his work, is well aware that fish diets can give anomalous results, but this new research on the variety and size of the error surprised him.

“I had not anticipated an error of up to 2000 years,” he said.

“The implications of this discovery are fairly frightening, because it is crucial for archaeology to have a reliable dating procedure.”

“An error of a few hundred years is acceptable when you date Palaeolithic finds, but an error of 2000 years is of great importance, even for the oldest periods.”

Riede highlighted the need to look at more reliable dates (for example Carbon-14 dates of short-lived terrestrial plants or twigs) and compare them to the now highly unreliable dates from cooking pots.

It is worth noting that even charcoal from a camp fire could be another error source – the “old wood effect”: where the charcoal dated might be from the innermost ring of a 500-year-old tree which was felled 100 years before it finally ended up in the camp fire. Knowing what might cause an error is vital when it comes to dating.

An added surprise

Now armed with the knowledge produced by The Aarhus AMS 14C Dating Centre, a group of researchers actually cooked some fish stew in ceramic pots.

The group, after making their own pots, boiled up the various ingredients – including a freshly caught fish with a Carbon-14 age of 700 years. They then succeeded in burning the meal onto the pot fabric, which was then taken for dating.

Even though the food crust was made only weeks before, Carbon-14 dating returned a 14th century date and thus provided evidence that food crusts on pottery take on the same age as the ingredients.

These results reveal that freshwater reservoir effects have to be seriously considered and understood whenever residues on prehistoric pottery is radiocarbon dated. The same also applies to the bones of humans who had eaten significant amounts of freshwater fish.

This is real food for thought.

Source: Past Horizons