The unbaked clay tablets found by Mr. Seton Lloyd and Bay Nuri Gökçe at Sultantepe were transported to the Archaeological Museum at Ankara in June, 1951. By a happy coincidence I arrived in Ankara shortly afterwards and was able to spend a period of about five weeks in examining the collection. At the beginning of my stay in Ankara the tablets were drying out well, and it was possible to brush off the accumulations of earth from many of them without damage. Others, however, resisted this superficial treatment and will have to be baked before an epigraphist can begin to work on them.

The questions which were uppermost in our minds at Ankara were the ancient name of the site, the date of the finds, and the general nature of the hoard of tablets found in a large pile in sounding F, since this in its turn would throw light on the nature of the building that contained them. It was obvious to me at first sight that these tablets closely resembled in script and appearance those from the library of Assurbanipal at Nineveh and that the majority of them contained religious or lexicographical texts. But since they were unbaked, it did not seem very probable that they had formed part of a royal library. My first attention was therefore directed towards the colophons, or scribal notes to be found at the end of many of the texts, since these were likely to contain the most direct clue to the nature and purpose of the collection and might also mention the date and place of writing.

No attempt has been made yet to collect the widely scattered information on the history and topography of Ḥarrān for the period from the Arab conquest in A.D. 639 until the destruction of the city by the Mongols in A.D. 1260. Such a compilation would form a useful sequel to the monograph of Adam Mez, which covers the period up to the Arab conquest and to Chwolsohn's work on the Sabians which, despite its venerable age, remains the standard work on the subject.

Several monographs on Ḥarrān were written during the Middle Ages but no Mss of them have so far been discovered. A good deal of material can be collected from historical and geographical texts. Of these none is more detailed or more worthy of attention than that of Ibn Shaddād. The account of this writer, available so far only in Ms, is contained in a chapter of his Survey of Syria and the Jazīra.

The site of the Cilician city known during the period of the Roman Empire as Caesarea, Caesarea by Anazarbus, or Anazarbus, is some 28 km. south of Kozan in the Turkish province of Seyhan. The place still retains its name as Anavarza, a hamlet built just outside the walls of the ancient city.

The choice of Anazarbus as a subject for research may seem to require some explanation. It was never one of the great cities of antiquity; indeed it is not yet certain that it even existed before the 1st century B.C. It never received more than scant notice from ancient authorities, while modern travellers, hampered by bad communications, a difficult climate, and a lawless population, were unable to spend more than a day or two at the site and consequently could not give more than a general idea of its monuments.

In 1949, as Scholar of the British Institute of Archaeology at Ankara, I had decided to study the Classical architecture of Cilicia, and at the same time to collect information on the many mediaeval castles in the district.