The stable isotope compositions of wool from sheep across different regions of Turkey have been determined to help identify the regions where ancient carpets were fabricated.

In the archaeological field, attributing the correct provenance to any object is one of the principal aims. Placing the recovered item in its correct time frame and geographical location are key factors in understanding the object, its use and the environment in which it was produced. The same accurate authentication is vital to prevent fraud by unscrupulous scientists, as evidenced earlier in 2005 when the disgraced German academic Reiner Protsch von Zieten was caught out. He had discovered a skull fragment which he claimed was 36,000 years old and was the famous "missing link" between Neantherdals and humans. In fact, it was only 7,500 years old.

One area that archaeologists have difficulty with involves carpets and other wool-based textiles. It might be fairly easy to attribute an ancient Turkish carpet to Turkey based on the design. The structure of the warp and weft might even help to determine its approximate age, since the manufacturing techniques have evolved with time. But how can you be certain that a carpet comes from a particular region of Turkey, or the Middle East, or Asia?

The situation is complicated by the fact that the carpet is one commodity that has been widely traded to all corners of the globe, so tracing a carpet back to its roots is a tricky task. Ancient goods were not bar coded.

However, scientists in the UK have been trying to establish the existence of an internal bar code in the wool from which Turkish carpets were made. They hope to base the code on the in-built stable isotopic signatures of the wool from sheep which graze in different regions of the country. Any characteristic isotopic signatures in the keratin from the wool would also show up in the corresponding carpets, confirming the area of manufacture.

Bradley Hull and Robert Hedges from the Research Laboratory for Archaeology and the History of Art at the University of Oxford, with Jonathon Thompson from the Department of Eastern Art at the Ashmolean Museum measured the C, N and S isotope ratios in wool taken from 13 herds across Western and Central Turkey, in the region known as Anatolia. The herds were all close to known carpet-producing sites.

The delta³⁴S, delta¹³C and delta¹⁵N values all increased generally from the west to the east of the region. The delta³⁴S values were in the range +3.5 to +9.5 permil and did not show the expected sea spray influence on herds grazing near the Mediterranean Sea. Published work on particulates in Israel, which has a similar weather pattern to Turkey, revealed an apparent distortion due to sulphate from sea spray blown inland. The researchers suggested that the absence may be linked to the way plants absorb S from their surroundings.

The delta¹³C isotope values ranged from -25 to -20 permil, within the normal range for animals that feed on the so-called C3 plants, whereas most of the delta¹⁵N values were in the range +4 to + 8 permil. However, although each of the 13 herds had unique isotopic signatures, there was insufficient variation overall to allow them to be clearly differentiated from one another.

Wool from two of the herds had surprisingly high delta¹⁵N values of +9 and +9.5 permil, well above the other nitrogen signatures. These were the only two herds that grazed in urban environments, where human influence was possible, as opposed to the other 11 herds that grazed in rural surroundings away from human habitation. The researchers speculated that the high values were caused by human and animal waste and decaying plant matter, which are rich in nitrogen and would enrich ¹⁵N in the soil and be transferred to the grazing plants.

This finding has major implications in the field of archaeology as a whole, although more urban samples will have to be tested for confirmation. "If animals kept in and around humans have an anthropogenically enriched signature due to changes in soil characteristics and plant N, and are then incorporated into human diets, we would see higher N in humans, and may make an assumption based on their diet due to the high N." Hull told spectroscopyNOW.com. "If the enrichment is simply due to keeping animals in proximity to human activity, then our understanding of diet through N stable isotopes would need to be adjusted."

For now, the team are extending their research to include hydrogen and deuterium isotopes, which may help to discriminate further between the sheep from different locations. They have also begun to examine carpets from known sources that are held in the Beattie collection at the Ashmolean Museum to compare their isotopic values with those of the wools.

Related links:

• Gallery Anatolia on Turkish carpets
• Research Laboratory for Archaeology and the History of Art, University of Oxford
• Ashmolean Museum
• The Beattie Collection at the Ashmolean Museum
• Rapid Commun. Mass Spectrom. 2005, 19, 3187: "Stable isotope variation in wool as a means to establish Turkish carpet provenance" 

Article by Steve Down