Frequently changing position of Caria lay in the political conjuncture that she had to witness numerous hostile powers of the ancient world. Seemingly, it was a land of aspiration, back to the 4th century B.C and thence. A sub-region, acknowledged as the Carian Chersonesos in the 5th century B.C, now the modern Bozburun Peninsula, also had to pull through various episodes in a time span of more than two centuries until the Romans troops arrived Asia Minor.
It seems that the Rhodians benefited from the “periods of political instability to expand their holdings on the mainland” as 280 B.C was a burst period[1] when the Diadochi were struggling to hold power over the territories Alexander the Great left behind. The Bozburun Peninsula, once being the focal point of the Classical Chersonesos, became a Hellenistic periphery (equipped with the demes) with the Rhodian takeover. It was, without doubt, the Rhodes’ successful diplomacy attacks and the ability to control many markets in the periphery during the political turmoils but more than that, the reason perhaps lay in the pre-established relations with the Carians at opposite side of the Island. The mainland was begun to be called as the Rhodian Peraea (hereinafter referred to as the “Peraea”) in the Hellenistic era. The introduction of a new deme system on the mainland or the replenishment of old Carian territorial forms brought the necessity to maintain a continuous alliance with the three old poleis (Ialysos, Lindos, Kamiros) and ended up serving the interests of the Island over time. We shall not question whether it was a top-down strategy imposed by the Rhodian State following her synoecism in 408 B.C or based on the consent of indigenous Peraean populations due to long-recognized amicable relations[2] before the Social War (357 B.C) or their Hellenisation in the social profile[3] . What is certain is that each “new” deme of the Peraea was assigned to one of those three old poleis. [4] As of the date of modern publications, we have idea about the relational status of the Peraean demes with the Island and their location on the mainland (e.g. Physcus (Marmaris) attached to Lindos or the Kamiran deme of Thysannos (Söğüt), however some of them still need reassessment (e.g. Cryassus restored to Ialysos)[5] including their physical positioning, and hopefully verification after a future scrutiny. Yet, an unlocated deme of Hygassos/Ygassos about which the discussions are centered hereunder, has proven very little. The knowledge disseminated by the ancient writers and scholars have led to confusions in fixing a thorough toponomical expression which go far as somewhere around Syrna (Bayır) or Kastabos nearby Hisarönü but already mark the presence of an ethnic origin- the Hygassians.[6] Occasionally linking with the surface material, some authors address the neighboring site of Losta- possible Sinus Schoenus, also associable with Hellenistic Selimiye. Not that far, Orhaniye and Turgut villages are time to time underscored to have had relation[7] with what we question.
Losta (Selimiye)/ Hygassos?
Despite the controversial views put forward by the scholars, we think that the territorium of Hygassos and the deme center is a little far from the given. As we deem, a search for the alternative loci should go further in the east of Selimiye. Before all and as per the practical and theoretical side, we may need to make a revisit to the ancient sites within the borders of the “modern” village, thus try to explore a hypothetical catchment area.
By tracing a number of ancient settlement remains in the downtown and taking into account now absent funerary inscriptions (previously reported for the 5th-3rd centuries B.C and that take the foremost part in comprehending the ancient context), we can say, the surrounding areas could have acted as the mini-hinterlands serving coastal Losta[8] which can be easily glimpsed in the north and partly the west of the village (Map 1). Around Losta, the abandoned terraces spoilt due to modern public works catch the eye with fertile soil cover (Fig. 1). An area suitable for settlement is situated between Gemecitdüzü Tepe, which is rich in respect of agricultural terraces in the inner west; and Gemecik- a wide enclave squeezed between Bahçeiçi Quarter and Akçakaya Tepe, lying in the coastal area. Gemecitdüzü Tepe is in need of attention with typical Peraean architectural blocks, few cisterns and wells and remaining walls of rural dwellings scattered over the plain area (Fig. 2). Physically, the site seems to have kept in contact with the neighboring inland site of Avlana District in Bozburun, (ancient Tymnos). Large (although modern) polygonal walls travelling the western slopes of Kelmusa Tepe facing the enclave suggest a suitable place for an early settlement in the vicinity, however, it is still difficult to assert a single period. Notwithstanding and regardless of period, this wall range recalls “Dema” walls which once divorced Athens and Eleusis but physically connected the two mountains in the 4th century B.C.[9] If now absent walls/ruins or any parts thereof were replaced by the recent works, the enclave encompassing Gemecitdüzü could have had relation to Losta or a physical link with the deme of Tymnos. We can never be sure at the moment unless links in the social profile are promoted with the Tymnians[10] or any other. Hence, we take an advantage of familiarizing the site with Losta. In spite of the abundance of characterless pieces, the potsherds are datable to the Hellenistic and Roman periods on a large scale and disclose various decorations on the rims. Rarely found dark tape rims and decorations on the body fragments seem to address the Roman era (Fig. 3-A,B). Looking at the type of masonry and some few coarse wares (particularly the amphorae handles and olpe fragments), it is worth considering that the site could also have been occupied during the pre-Hellenistic era. Relationally, the Carian presence is open to discussion since a cluster of rock-cut dwellings (facing a possible necropolis on its east) in the north of the mentioned wall range present a simple but compact hilltop setting. Nevertheless, the potsherds (though uncountable) attributable to this small compact settlement cannot be securely dated to give a basis for the chronology of the dwellings. An additional site in the northeast of Gemecitdüzü is Pınarçukuru Location which seems promising with a handful of ruins (having proximity to a small, natural spring) but extensive terracing appears to have disturbed the layout of the site which maintains a fair vision of coastal Hydas and part of the Cnidian Peninsula. Also, a coastal enclave falling to the east/northeast of Losta (Erler Location) has revealed evidence about a small size isolated farmstead (Fig.4-A) and ancient terrace cultivation around. Despite little surface material including some Roman sherds, the site must have been also occupied during the Hellenistic period as the Peraea has a reputation with rural landholdings engaged with agrarian way of living out in the chora. The farmstead disclosed some fine indicators of pressing activity (in the form of possible mola olearia[11]) (Fig.4-B), a cistern (lying nearby) and few stamped amphora handles (possibly Hellenistic) as well as quite disturbed walls and terrace relics situated in the middle of the agricultural enclave where a dried up stream ran.
Running an eye to the southeast of Losta, we can easily notice the narrow valley across which a temporary stream (Çaykuyu Dere) runs. The valley is physically interrupted with Karatepe and the lowlands of a high hill- Kaletepe in the north and south, respectively. It continues until a strait, which meets the borders of Kızılköy that is situated further in the east of Selimiye. A robust fortification (Fig.5-A,B) on top of Kaletepe is of attraction with its ramparts worked with polygonal masonry, and simple military barracks. It offers quite a high visibility, watching the open seas and directly facing the Cnidian Peninsula. On the road to Kızılköy, at the narrowest point of the valley where the strait appears, there is a small rock-cut conical shelter which could have served as a watch-post. Two ancient farmsteads[12] stand at lowest codes of the valley. Few amphora bases suggest the Hellenistic period. Some other buildings, which are situated at moderate distances from each other but completely disturbed today, could have formed a cluster of dwellings in the valley. The wall remains and the columns of the farmsteads are recognizable. The plan of the smaller Hellenistic farmstead is in a better condition whereas the larger one, which is a few minutes’ walk from the former, is mostly disturbed. This larger one suggests a sophisticated plan and lavish design along which an elite residence, oikos or a cultic edifice is subject to discussion. If so, the elite building could have served as a base, a controlling authority in the vicinity. Indeed, it has a safer position as it does not lie that far from the strait mentioned above. The building technique, its roundish plan equipped with wells and surrounding agricultural terraces make it more distinguished.
The northern sector of the valley and Çaykuyu Dere is interrupted with moderate elevations. Up on a series of hills (but mainly Karatepe), one can trace the ruins of dwelling clusters, presumably pre-Hellenistic. This isolated network of settlement seems rather early when compared to the ruins recorded along the valley, in respect of the masonry technique and positioning. We can barely report ceramic pieces but can say, many ruins catch the eye with polygonal walls. Proposing a core settlement, the site (pre-Classical?) on top of the steepest hill has high visibility. It reveals the traces of early water works (Fig.6) and overexploited agricultural traces behind. The boundary lines of the dwellings are noticeable from each direction. Watching the remote bays, the site strongly addresses security concerns, standing far-off the coastal area. Hence, further questions on the Carians’ presence might to be posed.
Lying inland, Kızılköy[13] is quite disturbed due to modern public works. We recorded numerous potsherds (suggesting the Hellenistic and Roman periods) as well as three undisturbed cisterns made of ashlar walls at the modern low code terrace fields. Reused blocks are traceable on the walls of a late construction in the center. It might be this construction or a neighboring one in close vicinity, on which a funerary inscription mentioning Leto was once (indirectly) reported[14] while some others were detected to have been built onto the window of a house. Not that far but further inland high above Kızılköy, there rises up a double-topped hill, known locally as Asarcık (in the west of Güncebaşı Tepe) associable with an Acropolis[15] at the peak. The slopes of the Acropolis draw attention with regularly dressed walls and large block scatters (often regarded as tombstones) (Fig.7-A,B,C). Along the valley falling to the east, a Hellenistic terrace worked with an elegant masonry technique appear (Fig.7-D) almost in an undisturbed position.[16] Benter, without addressing any coordinate, talks about two settlement areas in the inner parts of Selimiye. He speculates that the bay of Selimiye was used as the harbor of this settlement area.[17] Presumably, one of them, which he pinpoints as Asarcık, is the Acropolis.
The visibility is very high at the Acropolis (Fig.8-A). Fortification walls, partly appearing in mixed, irregular or pseudo-isodomic ashlar masonry make their course in the N-S direction (Fig.8-B). The enclosed area of the Upper fortification measures about 2 ha, its perimeter is 880 m. The physical appearance and positioning of the Acropolis is quite similar to those of some other Peraean demes observed during 2009- 2012[18] campaigns. The lower settlement lies in the east of the Acropolis, along the narrow valley facing Güncebaşı Tepe. A dried up water course divorces the valley into two up to the spot where more elegant constructions are visible. Typical Peraean blocks and water elements make up the bulk of ruins. Settlement terraces are both associable with dry rubble or polygonal masonry, and isodomic, bossaged walls. By looking at the rear façade of the sacred building affiliated with a Hellenistic inscription[19] (possibly a public edifice/ temple?- (Fig.7-D), it was quite recognizable that larger in-situ stepped blocks were used as the supporting architectural elements (Fig.9-A). The density of settlement increases at the upper codes of the valley where the abovementioned and below cited inscription (44x47 cm) dedicated to the Aphrodite cult and another inscription on which “А” sign, are visible (Fig.9-B,C). The sign might be a letter, which could have marked the gate number of a house. Behind the sacred building and its temenos wall, a rock-cut niche into which a sculpture was presumably placed, is now disturbed. The inscription (3rd- 2nd centuries) mentioning the Aphrodite cult[20] and noticeable with the reading Karneios[21] was detected in a smooth position during the 2010 campaign. The reading is as follows:
“[’A’φ]pοδίτάς Καρνείου
[ίκά] ή apva έριφον καί
[’Aγ]ριανίου έκται ίσταμέ-
[νο]υ βούν καί έρίφους δύο.” [22]
A side note might be that the month Karneios was attested on Rhodian amphorae with date/month of fabrication. This name was used twice with 1.57 % among 262 stamped amphorae collections found at Rhodes.[23] The word was also used on one of the stamps (275-220 B.C) recorded in Labraunda.[24]
A small pocket plain in the lower settlement divorces the Acropolis from the hill at the eastern opposite. Indeed, the majority of dwellings are orderly ranged beginning from the said plain. They are scattered across the moderate slopes, creating a compact settlement around the same spot. The entire settlement extends toward Yenisarnıçdüzlüğü location, in the northern direction via an ancient road. Over the extended area, additional ruins and potsherds (particularly Hellenistic and Roman coarse wares and the fragments of cooking pots, amphora bases) were observed (Fig.3- C,D). This part is accessible from the entrance of the Acropolis lying in the northeast where the column pieces (Fig.9-D) seem to approve the case. Brief to say, the ruins of dwellings are scattered across the eastern slopes of the Acropolis (Map 2) and the low code hill situated at the opposite side, however, dense vegetation makes it difficult to access many of them. Situated inland, the location of the Acropolis and the lower settlement must have acted as a natural shelter for the inhabitants. At a much lower code, lying in the modern fields near the main road, an ancient building, probably a farmstead is recognizable with in-situ walls and potsherds scatters. Presumably, this building fell into the catchment area of the Acropolis rising behind.
In the southeast of the Acropolis, at the opposite side of the modern road, a watch tower situated near Tülü Tepe may mark a strategic location. It could have been in charge of guarding the ancient borders, cross-cutting the grooved terrain. On the other hand, Kayalı Bay, which is reached via a stream (Kayalı Deresi), could have been an ancient route/ runaway corridor which also connected the Acropolis to the open sea. Lands in the environs of the Acropolis are quite fragmented, it is perhaps why many cisterns constructed with polygonal stone and opus quadratum may be found at regular intervals (e.g. the ruins of a farmstead with in-situ base walls and four cisterns found at the junction of Kızılköy-Bayır road). Here is a location between Hayıtlık and Eren Tepe where the alluvial lands were drilled for underground water. The route gives way to numerous pocket plains and agricultural terraces in the environs.
Concluding Remarks
Considering the debates arising from the epigraphical limits and any other, we opt to seek the deme of Hygassos out of the physical barriers of the demes/ sites noted by scholars, up to now, except the very case of Losta. Although evidence about the relations of the Peraea with Rhodes is more intact in terms of literary sources, coinage and epigraphy for the Hellenistic period, we may not technically want to be contented with the inadequate number of fragmentary pieces to seek out a way for a realistic location for Hygassos, however can feel safe about the presence of the associated ethnic. Being aware of the absence of satisfactory epigraphic evidence, the reason of our assignment is preferably owed to the presence of a yet uncertified Acropolis whose silhouette is quite well-defined with a compact plan and a possible catchment area extending far as Losta and perhaps ruling further in the east/southeast. Why the Acropolis also seems to be a promising site is that the inland positioning amongst the hilly topographies and exhibiting a compact design scattered over a limited topography makes it of value at the theoretical level, as the Hellenistic demes of the Peraea must have inherited many aspects from the Carian way of living. We are aware that the imprints of ancient sites stretching across the western/northwest Selimiye could not have necessarily had relation with an inland Acropolis or even with the immediate coast of Losta. But, as the topographical advantages seem to overweigh toward the west of the Acropolis, there is chance to state that the catchment area of Hygassos? could have extended across the coast and a little further. Kaletepe, maintaining a very strategic position could have guarded the limits of the deme (possibly acted as a garrison in the upcoming periods as well) in the southern sector and that such a positioning must have ensured the frontiers of the core site(s) whether this be the Acropolis or somewhere nearby Losta. We also recognize the fact that there is no alternative but to leave the floor to further studies which are expected to proceed with the subject matter at some time in the future.
FIGURES
