The previous oldest-known example of pottery was found in Japan, dated to an age between 16,000 and 17,000 years ago, but debate has raged in the
archaeological community as to whether pottery was first made in China or Japan.
The most recent dig at Yuchanyan was in 2005 by a team led by Elisabetta Boaretto of the Kimmel Center for Archaeological Science at the Weizmann
Institute of Science in Israel. They believe they have found a more precise way to read the history of human activity written in layers of sediment,
"The way people move around and mess up caves is very difficult to see archaeologically," David Cohen, an archaeologist at Boston University and a
co-author on the research, told BBC News.
"Imagine you have a fire and then people come in again have another fire and another, so you have the ashes of all these fires building up but at the
same time people are digging and clearing, pushing things to the side; this messes things up.
Fragments from a 1995 dig at Yuchanyan form a cauldron
"If you have an open-air site, you sometimes get a very clean 'layer cake' stratigraphy. Archaeologists before haven't looked at this closely
enough to realise what's going on in caves so they interpret this stratigraphy as a layer cake. But in actuality, it's 'lenses' of stuff that's
been mixed up and moved around."
It is comparatively easy to find evidence of human occupation in caves through the dating of charcoal from fires or bones from long-ago dinners, Dr
Cohen said. However, because of the unclear layering of sediment it is not easy to correlate well-dated layers with the pottery that may be nearby.
Part of the problem lies in the areas over which previous digs have searched: squares of perhaps five metres on a side.
"It's an issue of association, knowing where everything comes from in space across the cave," Dr Cohen explained. "If you're excavating in a huge
unit, you can only say it comes from within this 5m area and this 20cm of sediment, and that's not good enough for understanding human activity."
Instead, the team worked in sub-divisions of just a quarter of a metre square, painstakingly collecting bone and charcoal fragments. The samples were
then radiocarbon dated, revealing a clean distribution stretching between 14,000 and 21,000 years ago.
One fragment of pottery was found in a layer between two radiocarbon-dated fragments that both measured about 18,000 years old, taking the record for
The team hope that their smaller-scale searching and taking into account the effects of human activity on cave stratigraphy will help with future digs
at Yuchanyan, and elsewhere.
"It's a fantastic cave, and we hope that the way the way these excavations were done would set a precedent for how other caves will be looked at,"
said Dr Cohen.
Dr Tracey Lu, from an anthropologist at the Chinese University of Hong Kong, who was not an author on the latest study, noted that the dates reported
in this paper were slightly older than dates on pottery found in Japan.
However, she said the accuracy of radiocarbon dates in the limestone area has been under debate for many years.
"I agree that pottery was made by foragers in South China," she told the Associated Press news agency.
"But I also think pottery was produced more or less contemporaneously in several places in East Asia... from Russia, Japan to North and South China
by foragers living in different environments."