This was the second season of excavation at Amorium in east Phrygia, and the team worked for five weeks, from 24 July 1989. Our main aim is to trace archaeological changes and developments within the city from Hellenistic times into the Selcuk period. We carried out a general survey of the Upper Town by a regular 25-metre grid, and we also excavated three trenches, one in the Upper Town and two (which we started last year) in the Lower (Fig. 1). A preliminary analysis is underway of the pottery and small finds, which in the Late Roman and Byzantine periods appear mostly of local manufacture. There are some relevant excavation-sites in Asia Minor for study of the Late Roman period, but there is very little research geared to the so-called Dark Ages, especially inland. Amorium is a major site, virtually untouched, and the city offers a rare opportunity to examine an early Byzantine urban landscape. The excavation so far has been very successful, and has highlighted the site's great potential. Next year, we shall try to clarify the chronology, by more intensive excavation of the existing trenches.

This is the second publication of new fragments of Diogenes discovered by participants in the Oenoanda project of the British Institute of Archaeology at Ankara. Thirteen of the twenty-three fragments which came to light in 1974 have already been published (Smith H). The present work presents the remaining ten fragments (NF 55, 57, 66, 68, 70, 73, 81, 83, 92, 101) found in 1974, forty-four fragments discovered in 1975, and one fragment (NF 94) recorded in 1976. In 1975 we discovered three more fragments recorded in the nineteenth century (HK fr. 4, 23, 25) and a piece broken off the bottom of HK fr. 42.

It is a pleasure to acknowledge, with sincere gratitude, the debt which we owe to the Republic of Turkey for generously permitting us to continue our work at Oenoanda. We are extremely grateful to Bay Hikmet Gürçay and his colleagues in the Eski Eserler ve Müzeler Genel Müdürlüǧü for their help in securing the renewal of our permit and for their kind interest in our work. We also thank Bay Salih Kütük, Director of the Fethiye Museum, and the local officials in Muǧla, Fethiye, and Seki for their courteous attention and assistance.

Modern scientific excavation did not start in Turkey until the nineteen-thirties and it is therefore not surprising that Anatolian chronology has been somewhat neglected in the past.

Since the war, however, Anatolian archaeology has made rapid progress. New excavations, publication of old ones and systematic field-surveys have added a vast amount of new material, much of it stratified. These new results call for a reconsideration of Anatolian chronology, not as an adjunct to that of Syria and the Aegean, but in its own right. The recent excavations at Kültepe, near Kayseri, undertaken by T. and N. Özgüç for the Turkish Historical Society, have produced literary evidence for relating the Anatolian culture-sequences to those of Mesopotamia in and before the Hammurabi period. This new material has a direct bearing on the vexed problem of dating the Hammurabi period and the rise of the Hittite Old Kingdom and at last enables one to establish the chronology of the area, not only in relation to Mesopotamia, but also in absolute dates.

Less than two thousand years ago Western Cilicia, the Cilicia Aspera of the Romans, supported a large population. Settlements in the interior were comparatively few, for the rugged character of the country forbids agriculture on any considerable scale, and communications are exceedingly difficult. On the sea-coast, however, those bays and natural anchorages which had been first the ports, and later the refuge of the pirate fleets that terrorised the Mediterranean until Pompey's campaign of 67 B.C., were thriving centres of seaborne commerce. Among the most famous of these were Corycus and Elaeusa (later Sebaste).

Hittite geography and Hittite religion have for many years been among Oliver Gurney's principal interests. It therefore gives me the greatest pleasure that this small contribution in his honour deals with both these subjects, especially since the text with which it is mainly concerned is one which I first studied many years ago under his supervision. Like so many others, I owe him a debt of gratitude which can never be adequately repaid.

For many years now it has been clear that in any attempt to locate the site of Nerik the most important fact to be kept in mind is that the town was for a long period in the hands of the Gasga-peoples. Their capture of it was allegedly in the reign of Hantilis, and it was calculated that five hundred years passed before it was recovered by Hattusilis III. More probably however it fell into enemy hands in the reign of Arnuwandas I and his queen Asmunikkal, shortly after 1450 B.C. But whatever the exact details may be, it has to be accepted that any town subjected to an extended period of Gasgan occupation must have been situated to the north of Hattusas, in the direction of the area known in Classical times as Pontus. More recently there has been further evidence pointing to a similar conclusion.

The use of an absolute chronological framework based on tree-ring calibrated C-14 dates has been recently proposed by D. F. Easton in his attempt “… to come to grips with the crucial and difficult dating of Troy” (Easton 1976:146). Easton points out that unlike Tarsus, whose relative dating vis-a-vis Mesopotamia and Egypt is stable, Troy's relative chronology is not agreed upon and this “impinges not only on Anatolia, but on the Aegean and Bulgaria as well”.

In establishing his chronology Easton uses, in addition to “the normal comparative methods”, two sources: (a) radiocarbon dates which, after calibration, especially when using Suess's calibration curve, affect both relative and absolute dating, (b) his reassessed stratigraphy of the Bronze Age levels at Troy (Easton 1976; 1977).

Easton in his new chronological structure has not taken into consideration certain facts and opinions surrounding tree-ring calibrated radiocarbon dating. In view of the persisting controversy regarding this scientific dating method, it is premature, at least as far as Anatolia is concerned, to replace the relative dates derived from historical synchronisms with calibrated “absolute” C-14 dates.

Between 959 and 965, Byzantine forces, directed and most often led by Nikephoros Phokas, launched a series of successful campaigns of conquest, particularly in the east. Although little studied, these conquests are significant. The subject of this article, the conquest of the Cilician plain in 965, represents the culmination of a century of Byzantine pressure on the Arab border states, of which Tarsos was the last. Simultaneously, the elimination of the Hamdanid emirate of Aleppo as a threat to the Empire helped to lay the groundwork for further successes in Syria after 965. It is the contention of this article that these objectives were intertwined and that Nikephoros Phokas followed a concerted and structured strategy through this period, aimed primarily at the conquest of Cilicia. Other objectives, though dramatic, should thus be seen as subordinate to this underlying objective.

Following Gough, the late Roman church complex at Alahan in Isauria is usually described as a monastic site built by the emperor Zeno (AD 474–491). Recent reinterpretation of the site by Mango and Hill has suggested that it was not a monastery but a pilgrimage site. This paper argues that none of the physical evidence at the site gives good reason to think of it as built by Zeno. The usual lavish scale of imperial patronage is missing. Alahan was a rural church complex built in the late fifth or early sixth century, whose most important feature is its state of preservation.

Located in the central region of the Anatolian mainland, Konya has played an important role in east-west and north-south cultural interactions since prehistoric times. In order to investigate the cultural geography of this region from prehistoric times to the Classical period surveys and museum work have been carried out since 1987 (Bahar 1991; Bahar et al 1996). In the course of this work some observations have been made on the Iron Age, which is a problematic subject for the central Anatolian region as well as for Anatolia as a whole. During the Iron Age the grey pottery known as Phrygian ware occurs over a wide region from the basin of the Meander in the west into central Anatolia (Mellaart 1955: 117; Dupré 1983: 82; Summers 1994: 241-52). We have previously suggested that this ware should be renamed ‘inner-west Anatolian ware’ or ‘Luwian ware’ (Bahar et al 1996: 65-7). It is significant that this pottery is encountered especially around Sarayönü and Kadınhanı where Luwian peoples were intensively settled in the second millennium BC.

Situated on the outskirts of the modern town of Kadirli is the best preserved late Roman church in eastern Cilicia. Its characteristics of design and construction are of pivotal significance for addressing the development of early Christian architecture in the region, as it represents the archetype of a definable group of churches in the upper Cilician plain, which link the building traditions of Syria with those of Anatolia and the southern shores. As it stands today, this structure known as the Alacami displays over a millennium and a half of local history, the liturgy of three different faiths and even more transitional building phases. In 1949 a full measured survey of the building was conducted by Michael and Mary Gough who in the same year were working at nearby Anazarbus. The results were never published and remained lost until 1994 when work commenced on cataloguing the Gough archive. The following study presents the full details of the Gough's survey for the first time and provides a detailed discussion of the monument in context, charting its development from the Roman period to the present day.

On my first visit to Turkey, in 1955, I was kindly allowed to copy and photograph a number of dedications to the god Mên which are now housed in the Classical Museum, Konya ; they would have been brought from the hieron on Kara Kuyu hill, just outside the colony of Pisidian Antioch, probably after the excavations conducted by Sir W. M. Ramsay in 1912. (In his article “Studies in the Roman Province Galatia, II : Dedications at the Sanctuary of Colonia Caesarea”, JRS. VIII (1918), 107 ff., he refers by number to a list of inscriptions which he seems never to have published.) Eleven of these stones, with three uninscribed monuments, have since been published by Mr. E. N. Lane in the first part of his work on “A Re-study of the God Men”, Berytus XV (1964), 36 ff. Mr. Lane writes : “It [the unpublished material] was all found ostensibly in the course of the excavations which D. M. Robinson carried out in 1924 on the site of the forum of Antioch”.

The moment when the Ten Thousand sighted the Euxine is one of the most haunting scenes to come down to us from the ancient world. Retreating from Cunaxa near Babylon in 401 BC, Xenophon describes how the Greeks fought their way northwards across Kurdistan to scale the Pontic mountains, and reached the sea at the Greek city of Trapezus, already more than two centuries old. By linking Xenophon's famous account with Hadrian's inspection of his eastern frontier, their route across the mountains, and their triumphant viewpoint, can be determined with some certainty.

About 120 miles before the Greeks reached the Black Sea, the ruler of a large and prosperous city called Gymnias, probably the modern Bayburt, sent a guide to Xenophon.

The identification of Old Persian yakā- with Dalbergia sissoo Roxburgh was suggested by Ilya Gershevitch in a memorable article in 1957 after extensive investigations in the Bashākard mountains of northern Iranian Makran (east of L. 58, c. 100 miles north-east of Jask). The sissoo has been defined as a “large deciduous tree, bark grey, heartwood brown with darker veins, leaflets 3–5, broadly elliptical or ovate”, and K. A. Chowdhury has studied its distribution in northern India, starting from the Indus, spreading east to Assam, sticking to the sub-Himalayan tracts and usually going up to 1,000 m. above sea level. It also extends far into the plains along river banks. In Baluchistan it grows in the Suleiman range between the Indus and the Zhob rivers, having been noted as early as 1890 by J. H. Lace in his study of the vegetation of Baluchistan in the Wam Tangi forest situated in the limestone ranges in the Hurnai district, where it reached a height of 35 feet. Its distribution within Pakistan includes the districts of Dera Ghazi Khan, Dera Ismail Khan, Bannu, Hazara, and it can be seen in Peshawar growing on the University campus and in front of Dean's Hotel (Pl. XXIIIa). A specimen from Nepal was given to Sir Max Mallowan in 1973 and is illustrated here (Fig. 1).

In Anatolian Studies, Vol. XIII, 1963, there appeared I. McNeill's “The Metre of the Hittite Epic”, known here as MHE. Some implications of MHE are explored below with regard to sentence stress in Hittite.

In MHE Mr. McNeill shows that the Hittite Song of Ullikummi has a stress-based metre, and that it is built up of verses, each with four stresses and divided into two equal cola. This system accords with those of the Mesopotamian antecedents of the Hittite epic-mythological genre. MHE's prime interest is in the parallels between the Hittite and the Homeric epic forms, and in particular in the use made by each of set formulae such as are suited to oral composition.

A large part of MHE explores the variations found in the Hittite formulae. A basic formula is exemplified by Aas siunas memiskiwan dais “Ea began to speak to the gods”. Each word is stressed, and the clause fills a whole verse.

The origin of felt is unknown. It is in all probability the earliest form of textile making. Like most textiles, it has a low survival rate in archaeological conditions, so that specimens from the earlier millennia rarely come to light. The earliest documentation on it comes from China, c. 2300 B.C. but it is from the Central Asian Steppes that archaeologists have so far found the earliest examples of it. It has been produced there ever since. Sir Aurel Stein made numerous interesting discoveries of felt on his many journeys in Asia.

Through T'ang writers we first hear of felt in Tibet and that the Chinese were familiar with the famous felt industry of Persia. Herodotus gives accounts of its frequent use amongst the Persians and associated tribes. Strabo, for example, describes the felt caps of the Persians as “high turbans of felt” which protected them from the cold winters of Media. The Medes themselves can still be seen in the fifth century B.C. portrayed in their rounded felt caps in the reliefs of the Achaemenian palaces of Persepolis. (Pl. XlXa)

The 41 rock-cut tombs at Etenna provide a clear impression of this very early and widespread tradition for the region. They allow us to compare the Lycian, Pamphylian, Cilician and Pisidian rock-cut tomb architecture and understand connections and discontinuities between them. They also illustrate how features of such tombs are based on natural and technical factors, on local architectural traditions, or on the rock-cut architecture of neighbouring cultures. But basic similarities, such as the cutting of a tomb chamber in the rock, could occur without any influence from other cultural regions, for instance because of similar burial needs, similar natural materials or similar architectural knowledge. The influences between the rock-cut tombs of different regions and periods can be seen in the particular details, and their relation to local burial customs. Questions are asked, such as: if there were Lycian tombs in the Classical period, why were there no rock-cut tombs in Pamphylia in this period; why did the ‘dominant Lycian Classical culture’ not influence Pamphylia; and what were the tombs of the higher social classes of the societies of Classical Pamphylia?

Twenty years ago Laroche set forth in a definitive article the evidence for the goddess Kubaba as the ancient oriental prototype of the goddess transmitted to the classical world as Kubelē or Kubēbē. Little enough new information on Kubaba has appeared since then. However, in the course of writing the entry “Kubaba” for the Reallexikon der Assyriologie, I felt that a detailed examination of the passages of the Hieroglyphic Luwian texts which refer to this goddess would be desirable, especially in order to take into account recent advances in our understanding of that script and language. What follows then is transliterations and translations of all Hieroglyphic Luwian references to Kubaba (except for uninformative fragments) and elucidatory notes. This study supplies the texts to the relevant section of Laroche's masterly sketch.

With six rock-cut tombs of Lycian type, the region south of the city of Kibyra (in the southern part of Burdur province) is the northernmost area of their distribution. Five of these are published here for the first time. Comparison with tombs in Lycia proper suggests a date in the fourth century BC. On the basis of this date and of similar tombs found in the region of Oinoanda and in the Elmalı plain, it is argued that the existence of this type of rock-cut tomb in what was part of ancient Milyas testifies to the political and cultural expansion of Lycia under ‘king’ Perikle in the 370s BC.

The purpose of this paper is to suggest mechanisms pertaining to the foundation of a new city on the Kerkenes Dağ, in the highlands of central Anatolia in the mid-first millennium BC. Archaeological evidence that Kerkenes was a new foundation is discussed, after which its thoroughly Phrygian culture is outlined. The core of the paper discusses possible explanations for the unexpected appearance of this new capital. Possibilities include Phrygianisation or acculturation, centralisation of pre-existing Phrygian settlements east of the Kızılırmak, eastward expansion of the Kingdom of Phrygia and a large migration from central or western Phrygia. It is proposed that a single large migration provides the most plausible explanation for the founding of the city, for the display of its Phrygian-ness and perhaps also for its ultimate failure.

Excavators have put forward opposing interpretations of the architectural sequence at the Early Bronze Age site of Troy. C.W. Blegen suggested that freestanding ‘megaron’ houses determined the visual pattern of the earliest settlement, while M.O. Korfmann compared Troy I to the circular layout of the Early Bronze Age site at Demircihüyük (the ‘Anatolian settlement plan’). Post-excavation analysis of the archaeological record from the excavations (1987–1992) in ‘Schliemann's Trench’ at Troy suggests a modification of the views of both Blegen and Korfmann. The local building tradition at Troy I was characterised by flat-roofed, long-room dwellings arranged in rows and sharing lateral walls. This peculiar settlement layout, recurrent at coastal and inland sites in the western part of Anatolia during the early centuries of the third millennium BC but unfamiliar at this time in the neighbouring regions of central Asia Minor and the Aegean, can be described as an Anatolian ‘row-house’ style. The present study applies the concept of vernacular architecture to the architectural record of Troy I and other contemporary ‘row-house’ sites to elucidate the social and cultural factors that shaped the domestic architectural traditions of early third-millennium western Asia Minor.