During the archaeological survey carried out in 1972 in the Middle Black Sea Region by a team directed by Professor Dr. U. B. Alkım, Chairman of the Department of the Ancient Near Eastern Languages and Cultures of the Faculty of Letters of the University of Istanbul under the auspieces of the Turkish Historical Association, the Ministry of National Education and the University of Istanbul, it was decided to send a preliminary expedition party to the southern and western neighbouring sectors of the region. The purpose of this expedition was to gain a general idea about the topographic structure and the intensity of the ancient settlements or other cultural activity loci of the region, upon which the later investigations would be based. This survey was realised in two phases: the first covers the area between Samsun-Havza-Vezirköprü-Durağan, the second Durağan- Kargı-Osmancık-Gümüşhacıköy-Merzifon. We have carried out the second part. The results and finds of the main survey between Samsun- Bafra and Bafra-Alaçam and the first step of this investigation will be published by Professor Alkım. But, we would like to announce that the amount of the known mounds has been threefolded. It is hoped that the collected material will clarify the interrelations of the central and coastal cultures in the pre-Hittite and Hittite Anatolia.
The region we investigated (Map 1) has either not been visited or has been neglected in more general and wide-scoped research programmes by the previous authors (von der Osten 1936, Kokten 1952, Burney 1956). None of them mention any finds from this area. Although very poor from archaeological data, this area became the subject of many philological studies. It is claimed that the region was full of important settlements in the prehistoric and historic ages. We know in general, as far as the Hittite written sources inform us, that there existed three different political or administrative units on the Black Sea coast and in its hinterland. These are the lands of Kaška, Pala, and Azzi-Ḫayaša. Though their existence at the Black Sea coast is generally agreed upon, there are divers views on their exact localizations. There are no objections based upon logical argumentation against the localization of the Kaška Land north of Merzifon-Amasya line and between Sinop and Ordu (Goetze 1930: 24 f, 1957: 92 fnt. 23, 1957a: map: Garstang-Gurney 1959: map 1), if we leave aside the theories of Cornelius who extends the Kaškean land from Samsun to Malatya in the south-east (1958: 1, 1958a: 246, 1959: 106), and localizes Pala around Bayburt (1958a: 244), and Azzi-Ḫayaša between the lower courses of Kızılırmak and Yeşilırmak (1958: 2, 1958a: 237). Pala lied, for some authors, north of the Anatolian plateau around Paphlagonia (Goetze 1957a: 48, i960: 45), and for some others near the upper course of Kızılırmak east of the Hatti Land (Garstang-Gurney 1959: 30 f.) There is a general agreement between the theories on the localization of Azzi-Ḫayaša. It is sought approximately between Karasu and the Black Sea (Goetze 1930: 24 f, 1940: 25 f, 40 f). According to this geographical setting, the area which is the subject of this paper, seems to lie between the Hatti and the Kaška Lands (Map. 2). This means, that this region was the scene of the Hatti – Kaška struggle which lasted throughout the Hittite political history. In fact, as explained above, many place names which occur in the documents of various Hittite kings are being sought in this region, and the most important of these toponyms is the holy city of Nerik. Therefore, the centre of gravity in the orientation of our investigation constitutes the determination of the fact whether or not the archaeological and topographical evidences verify the theories on the localization of Nerik.
As it is known, Nerik was one of the greatest cult centres of the Hittites, the loss or gain of which during the Kaška – Hittite struggle was very important for both sides. Therefore, the localization of Nerik would contribute to the understanding of the Hittite as well as of the Kaškean political history and historical geography (for documents and events relating to Nerik, see Haas 1970: 5-14). Goetze (1957: 92-96) had pointed out that Nerik is to be sought in the area north-west of Ḫattušaš and north of Ḫakmiš = Amasya. Garstang and Gurney (1959: map 2) were inclined to localize Nerik 100 km as the crow flies south-east of Ḫattušaš on the northern branch of Kanaksu. The defectiveness of this idea was shown by Goetze (1960: 46). Güterbock (1961: 92), depending on a Hittite document (KUB XXXVI 89) which was not evaluated previously, tried a more precise localization of the city. On the reverse side of this fragment we read (lines 12-14) :
12 ÍD Maraššantaš-wa annallaza : ipattarmayan aršaš
13 DU-aš-ma-war-an wahnut nu-war-an DUTU-i DLIM-an aršanut URUNerik-war-an
14 manikuwan aršanut
We translate this passage:
12 The river Maraššantaš previously flew in another (:ipattarmayan) course
13 But the Weather God turned him and let him flow to the Sun God of the Gods (DUTU-i DINGIRLIM-an). He let him flow
14 near the city of Nerik.
An important and certain fact, which can be extracted from this fragment, is that Nerik was situated on or near the river Maraššanta alias Kızılırmak. It is evident that this geographical fact is tried to be explained by a divine command, apparently to exaggerate the religious significance of the city. Güterbock accepts the expression of this passage as an aethiological mythos, and believes that it explaines a turn of the course of Kızılırmak. He claims that Nerik has to be near a sharp bend on the course of Kızılırmak. A hint about the direction of this bend is also given in the text: the Weather God turned the river towards the Sun God. Sun generally means East (Güterbock also considers the opposite). But, in what direction the river flew previously, is concealed in the meaning of the gloss :ipattarmayan. A similar word (note the single and double -p-) occurs in KUB XXXV 45 ii 22 and 48, ii 15 (= Otten 1953: 46 and 49; Laroche 1959: 146-7) in the phrase kuiš-an šaḫḫaniššatta kuiš-an ippalarišatta. Laroche, though he does not give the meaning of the word in his dictionary of the Luwian language (1959: 52), renders this phrase as “celui qui l’a sali, celui qui l’a souillé” in the translations of the fragments which he added to the same book (1959: 146). The reason of his above translation is probably due to the co-existence of the word šaḫḫa- “to soil” (1959: 53). For the adverb ipatarma he does not give any meaning in his dictionary, but, in a previous article (1956: 423) he claims that this word designates one of the cardinal directions like DUTU-ı DLIM. It seems that Laroche accepts the difference in the meanings of these words. Güterbock, on the contrary, thinks that ipattarmayan and ippataršiatta are of the same origin in spite of the single and double -p- and tries to reconsile their meanings (1961: 93 fnt. 42). If ipattarmayan meant “west” in contrast to DUTU-ı DINGIRLIM, and if the phrase kuiš-an ippatarišatta meant “whoever soils him”, the original meaning of the stem ipatar- could be “down”. In that case “sundown”, which is equal to west, and the degrading meaning of “to soil” could be correlated easily with one another according to Güterbock. But he renders ipattarmayan as “astray” in the passage cited above (Haas 1970: 153 translates this word as “in another course” without any interpretation). In spite of the importance of the determination of the exact meaning of this word for the localization of Nerik, Güterbock claims that it will not change the fact that Nerik was near a sharp bend of Kızılırmak, and adds that the most striking bend Kızılırmak has on its whole course is at Kargı. Then, he studies the texts KUB V 1 and KUB XXII 25, in which occur names of several settlements in connection with Nerik, and identifies the mount Ḫaḫarwa -on the slopes of which Nerik seems to be situated- with the mount Adadağ near Kargı. Güterbock’s ideas, based upon a sound logic, find a vast agreement among scholars like von Schuler (1965: 19 ftn. 6), ten Cate (1967: 44-61 and map II), and Haas (1970: 5), and attempts to localize many other places are made depending on Nerik's identification with Kargı and Adadağ.
Thus, in these studies the area we have investigated seems to have many important settlements.
After the starting point of our expedition at Durağan the road follows the coast of Gükırmak. Descending further southward, one can observe that the territory becomes rougher. The road then reaches to the western coast of Kızılırmak and follows the river until Kargı. The slopes of the valley are very steep. At both sides of the river-bed occur occasionaly alluvial accumulations not higher than 150 cm above the water level, but generally the river flows through abrupt rocks not yielding a passage. The modern highway is an artificial road constructed by cutting the western slopes of the river valley. There were no remains of any settlements from Durağan to Kargı. Only two tumuli were found between the villages Edilli and Alpaşalı, south of Durağan. We investigated Kargı and its vicinity both from tha north and the south, the slopes of Adadağ, and went via Kâmilavlağı to the hills of Karaağaçdağı to be able to survey the region as completely as possible. In the triangle of Durağan-Kargı-Osmancık we could find only one mound, and that is Cintcpe III (for Cintepe I and II, sec Professor Alkim’s forthcoming article). Cintepe is situated io km west of Kargı and approximately i .5 km south of the village Yeşilköy (Runkuş) on the Osmancik-Kargt highway about 400 m north of the junction point of the Devrcs Çay and the Kızılırmak. The hills of Adadağı and Kösdağı are on the south of the mound, from the top of which both the junction of Devres and Kızılırmak and the town Kargı can be seen. The Ilgaz mountain range lies ca 1.5 km south of Cintepe (Map 1). The mound is elongated in shape. On its northern and western slopes natural rock formations can be seen. There are no architectural remains on its surface. Potsherds were not abundant on the surface of the mound, which is ca. 5 m high. Most of the material was collected from the pits of illegal excavations carried out by treasure seekers (Figs. 1, 2, 3).
The surface material collected bears characteristics which dates it to the Early Bronze I and EB II and III periods. The pottery of the EB I from Cintepe can be divided into two groups: 1 - Black coloured and roughly burnished sherds. 2 - Brown and unburnished pieces. The sherds belonging to the first group arc not well-fired, and their cores arc either red or reddish brown in colour. They are straw and stone tempered. This group can be classified into two subgroups : a - Sherds of pithoi and kitchen ware, b - Sherds of thinwalled vessels. Thin-walled sherds have brown or buff patches on rims and/or on shoulders because of uneven firing. This type of pottery is very similar to the so-called Alişar Chalcolithic Ware which in reality belongs to the first phase of the EBA. The same pottery frequently and abundantly occurs in the Samsun region. The brown coloured unburnished pottery of our secound group is also badly fired and straw tempered. They are of inferior quality than those of the first group. They too show unevenness of colour on their surfaces. The potsherds we date to EB II-III have a better surface-treatment and are rather well burnished. Like the EB I material they too are hand-made. Their cores arc red like their surfaces. Their temper consists of fine sand (Fig. 4).
In the continuation of the area we investigated -that is the territory between Osmancık and Gümüşhacıköy- we could not find any settlements. We must admit that we have not penetrated into the interiors of this region. But, between the townships of Gümüş and Hamamözü further south, at the place called Kışla Mevkii, some 300 m south of the road, we found a small settlement, probably founded on a natural hill which is ca 25 m in height. The earliest pottery fragments encountered on the surface date to the Iron Age (Phyrigian). It should be pointed out that the location of this settlement was on one of the secondary natural roads connecting Central Anatolia to the Middle Black Sea Coast (Map 1). Approaching to Hacıköy, the road follows again the river valley, and 19 km before Hacıköy, it runs through a rough defile for approximately 3 km. Between Gümüşhacıköy and the limit of our survey Merzifon no archaeological remains could be detected.
As may well be seen from the foregoing explanations, the region where Nerik and other cities in connection with it are localized does not support these theories with its topographical and archaeological evidences. There are no loci of cultural activity other than Cintepe III in the triangle Kargı-Durağan-Osmancık. No traces either of the famous city of Nerik nor of the neighbouring settlements can be found on the slopes of Adadağ which is thought to be the Ḫaḫarwa Mountain. According to the results of this survey the theories on the localization of Nerik based upon philological evidence should be reconsidered under the light of the archaeological data, and other areas should be taken into consideration where archaeological and geographical facts are appropriate. If we interpret from this point of view the document KUB XXXVI 89 rev 12-14, which gives the best information on the localization of Nerik, the following results can be obtained:
1 — The text may not necessarily mean an actual change in the course of the river Maraššanta. As cited above, the author of the text could have exagerated the importance of the city by emphasizing the nearness of the river through a mythos.
2 — By a variation in the course of the river only a small turn or an unimportant change of the river-bed may have been meant in the text.
3 — Nerik does not have to be situated just at a sharp bend of the river, for there is no such remark in the text. It can be anywhere between the bend and the mouth of the river.
4 — If the phrase containing the gloss :ipattarmayan meant that the river flew in another direction and not towards west, then we
also have to consider the sharp turn the river makes at Durağan, where it abruptly changes its course from north-west to east, in connection w ith the possible localization of Nerik.
We learn from KUB V I that there were many settlements on the slopes of Ḫaḫarwa and in the vicinity of Nerik like Ḫanḫana, Tanzilia, Aštigurka, Talmaliya, Ziḫḫana, Iupapaena (Güterbock 1961: 94). This clearly shows that Nerik was in an intensively settled area. The Hittite sources frequently mention the “people of Mount Ḫaḫarwa", a part of which was at “peace” and another apparently in struggle with the Hittites. That such a big population could not find shelter on the hills of a rather small mountain was also accepted by Güterbock. Therefore, he called Ḫaḫarwa not a single mountain, but, a mountain range (1961: 94). Since it is not a mountain range, and since no settlements could be detected during our survey, Adadağ seems not to be an appropriate equation for Ḫaḫarwa. Even if we accept that all of the above enumerated towns were open settlements, it would not be logical to believe, that they have been eradicated without leaving any traces behind. Therefore, the theories forwarded, according to which Nerik was to be sought in this region, are opposed by the territorial conditions and archaeological data. As for the localization of other places in the same geographical limits, ten Cate, depending on the knowledge extracted from the written sources, equates (1967: 57 map II) Devrcs Çay (class. Blaëne) with Pala, Gökırmak (class. Amnias) with the River Daḫara (see also Güterbock 1961: 95), the Ilgaz Mountains (class. Olgassys) with the mount Kaššu, localizes Mt. Elluriya south of Gökırmak near Elekdağ, and the towns and townships Ziḫḫana, Šapidduwa, Atḫulišša and Tummana between the rivers cited above. He further places Timmuḫala near Vezirköprü, Durmitta near Havza, and Tapapanuwa between Boyabat and Durağan (Map 3). All these attempts at localizations are based on the connections of the places, using partly the quasi certain localization of Nerik as the key-point. But, historical events as well as the geographical and topographical conditions do not seem to be in favor of these theories. All these places, except Nerik, have been conquered by the Hittite armies during the north-west campaigns of Muršiliš II. How Nerik has been left untouched, when all the surrounding towns had to obey the Hittite dominance, is not an easily comprehensible matter, since the object of the military operations seems to be the conquest of Nerik herself (von Schuler indirectly touches on this matter in 1965: 57). Muršiliš reports in his annals (KUB XIX 37 iii 1 ff) that he directed campaigns to the neighbourhood of Nerik, but does not mention the city itself. Yet, we learn from the records of Tudḫaliyaš IV. (KUB XXXI 14/Haas 1970: 8 fnt. 5) and Ḫattušiliš III. (KUB XXI 8 ii 1/ Haas 1970: 9, 10 fnt. 1 and 2) that Muršiliš and Muwattališ had devastated the fields around Nerik without any harm done to the people of the town during their campaigns. The correctness of these expressions are doubtful: to leave such an important cult centre to the enemy after conquering the surrounding territory several times with shorth intervals cannot be true. We think that Hattušiliš III. and Tudḫaliyaš IV. wanted to honour their ancestors, who probably approached Nerik but could not be victorious. Some authors, Haas for example (1970: 11 fnt. 1), believe that the above-mentioned Hittite kings could not destroy Nerik, since the population fled to the mountains. Militarily speaking, the subjection of an evacuated town is easier than a populated one.
Until now we tried to demonstrate the irrelevance of the proposed theories on the localization of Nerik. We would like to dwell on the possibilities of a more appropriate localization. Therefore, we have to take into consideration the following conditions:
1 — Nerik must be cut off from the reach of the Hittite forces during the campaigns of Muršiliš II.
2 — It must be situated near Maraššanta.
3 — The territory where Nerik is located must be mountainous.
4 — Nerik must be in an intensively settled area.
Thus, we believe, that Nerik has to be sought north of the line Boyabat- Durağan-Vezirköprü, and between Kızılırmak and the western continuation of the Isfendiyar mountain range. This territory is cut by deep valleys, and is sharply undulated. The mounds, as far as we could detect during the main survey, show the density of the settlements. In fact, there are about 20 mounds, on the hills on both sides of Kızılırmak in the region between Bafra and north of Vezirköprü, some of which bear the characteristics of big towns with fortifications. On almost all of them, there are pottery forms of the first and second halves of the second millennium. To determine which one of the mounds could be Nerik is very difficult. We need more data on its location, and we have to wait for more written sources from the archives of Boğazköy to come to light. But if we accept, that Nerik was in this region, we can understand why it has not been conquered by Muršiliš while Daḫara and Tapapanuwa have been twice occupied by his armies at short intervals. This northern location also sheds light on the question why such an important religious centre has been lost to the Kaškeans: this region is connected to Central Anatolia by two natural roads. The first runs in the direction Çorum-Merzifon Havza-Kavak-Samsun, which is the course of the modern highway, the second following the route Çorum-Merzifon-Havza-Vezirköprü- Oymaağaç (that is a mound with remains of a city wall and a postern)- Alaçam (Map 1). Without getting hold of Vezirköprü, which is the gate to the northern regions, no military operations directed from Central Anatolia arc possible in the Middle Black Sea Region. This is why this region, which in fact was a northern protuberance of Hatti Proper, fell under the Kaškean domination when the military power of the Hittite Empire diminished.
CORNELIUS, F.
1958 “Zur hethitischen Geographie: die Nachbarn des Hethiter-reiches”. RHA 62: 1-17
1959 a “Geographic des Hethitcrrciches”. OrNs 27: 225-251
1960 “Der Text des HattusiliS III., geographisch erläutert”. RHA 65: 104-116
GARSTANG, J. ve GURNEY, O. R. :
1959 The Geography of the Hittite Empire. The British Institute of Archaeology at Ankara, London.
GOETZE, A.
1930 “Bemerkungen zu dem hethitischen Text AO 9608 des Louvre”. RHA x: 18-30.
1940 Kizzuwatna and the Problem of Hittite Geography. Yale Oriental Series, Researches XXII, New Haven.
1957 “The Roads of Northern Cappadocia in Hittite Times”. RHA 61: 91-103
1957 a Kleinasien. Handbuch der Altertumswissenschaft. C. H. Beck’sche Verlagsbuchandlung, München.
1960 Review of Garstang-Gurney “The Georgraphy of the Hittite Empire”. FCS 14: 43-48
GÜTERBOCK, H. G.
1961 “The North-Central Area of Hittite Anatolia”. J KES 20: 85-97
HAAS, V.
1970 Der Kult von Nerik. (Studia Pohl 4), Päpstliches Bibelinstitut, Rom.
LAROCHE, E.
1956 Besprechung von H. Otten “KUB XXXVI”. OLZ 51 · 420-424
1959 Dictionnaire de la langue louvite. Librairie Maisonneuve, Paris.
OTTEN, H.
1953 Luvische Texte in Umschrift. Akademie-Verlag, Berlin.
ten CATE, Ph. H. J. H.
1967 “Mursilis’ North-Western Campaigns - A Commentary”. Anatolie a 1: 44-61
von SCHULER, E.
1965 Die Kaikäer. Walter de Gruyter, Berlin.