Resizing Phrygia: Migration, State and Kingdom

Altorientalische Forschungen 2023; 50(1): 107–128 Geoffrey D. Summers Resizing Phrygia: Migration, State and Kingdom Dedicated to the memory of Oscar White Muscarella whose passion for Phrygia and things Phrygian was an inspiration https://doi.org/10.1515/aofo-2023-0009 Abstract: This paper considers the concept of Phrygia, its origins at the cusp of the Early Iron Age, the beginning of the process of state formation marking the start of the Middle Iron Age and the possible development of a large unified kingdom under Midas son of Gordias. Secondly, the paper turns to consider the aftermath of the Phrygian state following the death of Midas and the period of Lydian hegemony together with the establishment of a new Phrygian petty kingdom east of the Kızılırmak (the Red River, the classical Halys River) in the first half of the 6th century BCE. In order to assess the extent of the territory of Phrygia, it has been necessary to consider evidence for neighbouring polities, particularly the Neo-Hittite kingdoms to the south and ‘Cappadocia’ to the southeast and east. Two maps graphically illustrate the principal suggestions. Keywords: Phrygia, Neo-Hittite, Anatolia, migration, state Introduction The Kingdom of Phrygia reached its apogee under its most famous king, Midas, who ruled from about 740 BCE into the early seventh century BCE.1 We have no contemporaneous written historical sources from Phrygia itself, although there are a good number of cultic inscriptions which, together with a few alphabetic graffiti, are sufficient to indicate the existence of a literate class using written texts on perishable materials.2 Although the wealth of Midas was legendary and he is known to have led Anatolian coalitions against Assyria, his riches and strength may both have been inflated. While little by way of associated precious metals has been found, control of extensive labour is evident in the huge building works undertaken at Gordion, the Greek name by which his capital is known, as well as the enormous tumulus (Tum. MM) probably constructed by Midas for his father.3 Inauguration of new and wide-ranging research at Gordion under the direction of Brian Rose has included remote sensing and large-scale excavation together with a most welcome publication program. When the new campaigns of fieldwork began these re-evaluations were already underway, not the least important of which being a redating of the famous Destruction Level excavated by Rodney Young, raising it by around a century from circa 700 to 800 BCE.4 This redating, together with re-examination of the Destruction Level itself has disassociated the fire from any Cimmerian attack and has also demonstrated that not all of the citadel was burnt. It 1 For multiple kings named Midas and the chronology of Midas the Great see Berndt-Ersöz (2008). 2 Obrador-Cursach 2020. The only possible exceptions to the lack of epigraphy containing historical information are the Black Stone of Tyana (Tuwana), T-01 to T-03, (Obrador-Cursach 2020: 505–507 with references) and fragments from a single block, K-01, from the Monumental Entrance to the Palatial Complex at Kerkenes (Brixhe/Summers 2006; Summers 2022; Obrador-Cursach 2020: 508 where the findspot is wrongly given). Neither inscription is well understood, with any historical information contained therein likely to be secondary to a dedicatory or cultic purpose. One may add to this graffiti on a beam in Tumulus MM at Gordion, possible historical connections of which are discussed by Maya Vassileva (2021). 3 Recent overviews of Gordion in the Iron Age include: Rose (ed.) 2012; Rose 2017; 2021; Rose/Darbyshire (ed.) 2016; Voigt 2013. 4 Rose/Darbyshire (ed.) 2012. Geoffrey D. Summers, ENSA Nantes (Mauritius), E-Mail: [email protected] 108 Altorientalische Forschungen 2023; 50(1) is probable that the fire, destructive though it was, began accidentally.5 More importantly, perhaps, the work of Keith DeVries has shown that the massive program of raising and rebuilding the citadel was already well underway before the fire.6 The single most important result of the redating was to make the dynasty of Midas son of Gordias responsible for the heightening and extension of the citadel.7 This re-evaluation resolved the awkward problem of explaining how such an extraordinary program of construction could have been inaugurated following, so later Greek sources claim, the suicide of Midas by drinking bull’s blood after his defeat by Cimmerian hordes.8 Exciting results of the new investigations have provided an enormous amount of fresh information, including Ben Marsh’s documentation of the shifting course of the Sakarya River from one side of the city to the other as well as his identification of the Middle Phrygian Outer Town defensive circuit.9 Thus, the extent of wealth and power of Gordion and its greatest king, Midas, in the later eighth and early seventh centuries BCE have been brought into sharper focus.10 Over and above these developments at Gordion are discoveries central to an understanding of the placement of Phrygia within the wider context of the eastern Mediterranean and the Ancient Near East. These new resources include the extraordinary Anatolian Hieroglyphic, Luwian, inscription from Türkmen-Karahöyük (TKH 1) in the Konya plain together with archaeological evidence from regional survey of the plain that may, as discussed below, shed new light on the arrival of the first Balkan immigrants, or proto-Phrygians, in Central Anatolia.11 Also now available are two final reports on excavations at Kerkenes Dağ, a sixth century BCE capital to the east of the Kızılırmak that displays many aspects of Phrygian culture.12 It is not the purpose of this paper to summarise or evaluate in detail these results from recent investigations. Rather, my aim is to take a broader brush to examination of such scant evidence as there might be that can be related to the origins of Phrygia, the nature of the Phrygian polity, whether Phrygia may have been a single unified kingdom under Midas and what happened after Midas. There have been two opposing views of the Phrygian kingdom. On the one hand there is the commonly held assumption that Phrygia was a single kingdom bound together by aspects of common culture that included language, religion and material culture, ruled over by a Near Eastern-style despotic monarch, namely Midas. A more minimal view, possibly more widely held than expressed, sees Phrygia composed of smaller polities, perhaps resembling Neo-Hittite and Greek city states, exhibiting common cultural elements side by side with more local cultural and historical trajectories. In this view there would have been under Midas a coming together in the face of external aggression, be it from Assyria, Lydia or the more shadowy Cimmerians.13 Late Bronze Age Collapse and the Phrygian Migration In the later part of the second millennium BCE the Hittite Empire collapsed in the midst of a wider series of catastrophic events that included the end of Mycenaean civilisation and disruption across the entire Eastern Mediterranean. The extent, timing and causes of this collapse, which included what became known as the Tro- 5 For the accidental fire see DeVries (1990), Voigt (2012). Osborne (2020: 22) suggests a connection between the Destruction Level at Gordion and the defeat of the Land of Muška by Kartapu in the Türkmen-Karahöyük inscription (see further below). 6 DeVries 1990. 7 There were several kings named or given the title of Midas, the earliest of whom was said by the Greeks to have been a son of Gordias, mythical founder of the Phrygian capital from who in Greek tradition it took its name, Gordion. It is this Midas who, by almost unanimous consensus, is referred to as Mitā of Muški in eighth century Neo-Assyrian inscriptions. The Phrygian name of Gordion is unknown. In this paper Midas refers to Midas son of Gordias (Midas the Great) unless otherwise stated. For the names and chronology of Midas see Berndt-Ersöz (2008) with references to earlier discussion. 8 For one view on the myth see Berndt-Ersöz (2019). 9 For geomorphology of the flood plain and the course of the Sakarya River see Marsh (1999); for identification of the Outer Town enclosure as well as Outer Town surface survey and excavation see Voigt (2013). 10 For the suggestion that Gord might be equated with Slavonic gorod or grad to which was added the Greek diminutive ending –ion see Sotiroff (1969: 9). I thank Peter Kuniholme for bringing this to my attention. For a fuller discussion see Obrador-Cursach (2019). 11 Amongst an already considerable literature see Goedegebuure et al. (2020); Massa et al. (2020); Osborne et al. (2020); Hawkins/Weeden (2021); Massa/Osborne (2022). 12 Summers 2021; 2022. 13 The minimal view has been well set out by Genz (2009) and van Dongen (2014). Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 109 jan War in later traditions as well as the invasions of the so-called Sea Peoples, continues to generate intense scholarly debate that falls beyond the remit of this paper. However, recent overviews tend to ignore the western portion of the Anatolian Plateau, perhaps reflecting the relatively sparse evidence available.14 Before looking at the archaeological and then the epigraphic evidence it is pertinent to make some more general comments with regard to collapse, population and migration. The biggest limitation is, of course, the paucity of archaeological evidence from Phrygia. A great deal is known about the Early and Middle Phrygian levels at Midas’ citadel in the capital of Gordion that has been extensively excavated, if poorly published.15 Iron Age levels at Gordion date from the twelfth century to the Persian invasion of around 550 BCE (Table 1), with most of our knowledge of the settlement pertaining to the so-called ‘Destruction Level’ on the citadel, which was caused by an extensive, seemingly accidental fire now dated to circa 800 BCE (see above). The conflagration debris were sealed by a layer of clay up to four metres deep, probably coming from canalisation of the Sakarya River. At the same time, the gate was filled with stone that thereby preserved its walls to a height of no less than 10m. On the other hand, there was much less exposure of the Early Iron Age (EIA) and the Late Bronze Age (LBA) levels that have been explored only over restricted areas.16 West of Gordion, excavations at Midas City in the Phrygian Highlands did not reveal LBA or EIA levels, its importance occurring during the Middle Iron Age (MIA) and to a lesser extent the Achaemenid period.17 Of great potential is the site of Şarhöyük, ancient Dorylaion on the outskirts of the modern city of Eskişehir, where Muhibbe Darga and then Taciser Sivas excavated very difficult Phrygian levels beneath substantial later occupation.18 In the opposite direction, only 22km to the east of Gordion, lies Hacıtuğrul Tepe where unpublished excavations by Ankara University revealed evidence of Middle and Early Phrygian (using Gordion terminology) occupation including strong defences, but with no reported indication of a lower town. More recent survey identified EIA sherds. We shall return to Hacıtuğrul later. Of Phrygian Ankara little is known apart from the large tumuli, the contents of which indicate significant status. This all-too-brief survey demonstrates that, archaeologically, our knowledge of the EIA in Phrygia is very largely restricted to Gordion itself. Table 1: Terminology in use at Gordion with approximate dates. YHSS = Yassıhöyük Stratigraphic Sequence, Yassıhöyük being the modern Turkish name of the site. Adapted from Kealhofer et al. (in press). YHSS PERIOD DATES BCE 3B Early Hellenistic 330‒250 NOTE 4 Late Phrygian 550‒330 Achaemenid 5 Middle Phrygian 800‒540 c. 600‒540 Lydian 6 Destruction Level 800 (–825) citadel fire 6DL Early Phrygian 900‒800 first monumental building & tumuli 7 Early Iron Age 1150‒900 ‘Proto Phrygian’, Dark Age 8–9 Late Bronze Age III 1400‒1150 Later Hittite Empire There is general agreement that in Central Anatolia the LBA came to a sudden end and the Hittite cities at the core of the empire were burnt.19 At the capital, Ḫattuša, there is evidence of abandonment before the conflagra- 14 Recent overviews include Cline (2021) and Knapp (2021). 15 There are no final excavation reports on any of the extensive excavations of the citadel at Gordion. Most laudably Brian Rose, the current director of the Gordion project, is attempting to address this lamentable state of affairs. 16 Voigt 2013 with references. Voigt/Kealhofer (in press) provides detailed description of the LBA and EIA levels; Kealhofer et al. (2019) sets out the new radiocarbon evidence for the LBA/EIA transition. 17 Gabriel 1952; 1965; Haspels 1951; 1971. The Achaemenid period occupation of Midas City amounted to no more than a village. 18 The untimely death of Taciser Sivas has precluded comprehensive publication. More recent excavations by the Eskişehir Museum appear to have focused on later periods. 19 There is growing evidence of earlier destructions as well as of decline before the final collapse, but that does not alter the view that at least within the bend of the Kızılırmak the Hittite Empire collapsed about 1180 BCE. Recent overviews can be found in Weeden/Ullmann (ed., 2017). 110 Altorientalische Forschungen 2023; 50(1) tion, with the likelihood that much of the population living within the bend of the Kızılırmak moved to the southeast, eventually making the vice-regal city of Karkamiš a new centre of power. Of importance for the present discussion is that no city on the central plateau demonstrates urban continuity from the LBA in to the EIA, in contrast to the southeast where there is growing evidence for continuity not only at Karkamiš but also on the Euphrates at Arslantepe, ancient Melid, the latter being surprisingly small in relation to its apparent historical importance.20 Recent attempts to demonstrate continuity from the LBA into the EIA, including redating eighth century Hieroglyphic Luwian inscriptions back into the 9th century, have failed to convince in the absence of supporting data from excavation or survey.21 We are, then, back to a ‘Dark Age’ which, even if some light is now being shed on it, remains dark in the original sense of the characterisation – devoid of urban civilisation and literacy.22 That there were pockets where knowledge of Luwian Hieroglyphic writing was kept alive on the Anatolian plateau, or where real or imagined descent from Hittite dynasties was perpetuated, does not alter this picture. As yet, on the central plateau there is no evidence for what have been termed Hittite ‘rump states’ in Southeast Anatolia and North Syria. Future excavation at sites like Türkmen-Karahöyük have the potential to alter this picture. With regard to the area that became Iron Age Phrygia, we are then left with a picture of deserted and destroyed urban centres, admittedly evinced only at Gordion, and some level of regional depopulation. Lisa Kealhofer provides evidence for low EIA settlement density around Gordion where very few sites of this period were recorded, with about 70 % of ‘Phrygian’ sites being newly settled.23 It is not to be imagined that the entire upland area of western central Anatolia was totally depopulated, but it can certainly be envisaged that mixed farming in areas where rain-fed agriculture, never more than marginal, was a factor not only in the collapse of urban centres but also in the desertion of rural areas. Central Anatolia is today characterised as dwarf-shrubland or steppe and was the driest part of the Anatolian plateau throughout the Late Holocene.24 It seems impossible to deny that one major cause of the LBA collapse was severe drought combined with other socio-economic and political factors.25 There is modern documented evidence that a few short years without rain on the plateau could be devastating. In 1874 in the Province of Ankara, District of Keskin, a drought occurred of such devastating proportions that 81 % of the cattle and 97 % of the sheep died. Of the population of 52,000, some 7,000 moved out of the district and 20,000 died (ChristiansenWeniger and Tosun 1939). The traveller C. Naumann reported that in the Provinces of Kastamonu, Ankara, and Kayseri 150,000 people and 100,000 head of livestock (40 % of all herds) died. Hunger and sickness through the 1873–1874 winter killed another 100,000 people (Naumann 1893).26 Peter Kuniholm presents the tree-ring data for those years and cautiously discusses the difficulty of identifying such short-term catastrophic droughts in the dendrochronological record. The evidence just cited is not what climatologists term Rapid Climate Change (RCC), which they measure in decades or centuries, but a short and very sharp climatic ‘event’ for which environmental evidence will be very hard to identify in the archaeological or geomorphological record. It is easily seen that the area of central Phrygia would more readily have lent itself to grazing large flocks of sheep and goat than to rainfed cultivation. In the Hellenistic period the temple estate of Pessinus derived the greater part of its wealth from wool while in modern times sheep and goat were the basis of the economy around Gordion.27 One could therefore posit a scenario in which a severe drought led to a demise 20 This is a necessarily over-simplified summary; for a fuller picture of the current state of knowledge see contributions to Weeden/Ullmann (ed., 2017). For Melid see Manuelli et al. (2021). Osborne (2020) provides an overview of similar issues in the southeast. 21 D’Alfonso 2019; 2020: discovery of the TKH inscription now rules out the redating of the TOPADA inscription, see Hawkins/Weeden (2021). 22 Hawkins 2002; Summers 2017. 23 Kealhofer 2005. 24 Kuzucuoğlu 2015. 25 The literature is vast and inconclusive: for the seminal paper see Carpenter (1966), but see Kuzucuoğlu (2015) for a nuanced view. See also Knapp/Manning (2016). 26 Kuniholm (1990: 650) who kindly tells me that the original publication was Gassner/Christansen-Weniger (1942); see also Müller-Karpe (2009: 255–256 with fig. 4). 27 For zooarchaeological evidence for the importance of caprids at Pessinus and Gordion see Slim/Çakırlar (2022). For Gordion see Gürsan-Salzmann (2005). Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 111 in settled mixed farming in favour of pastoralism that included elements of seasonal transhumance. This, together with cultural factors associated with the Hittite collapse, could have induced lower populations. The picture that then emerges is one of a region between Sivrihisar and the Kızılırmak, including the district of Haymana, which would have been attractive and highly suitable for the settlement of pastoral groups who had migrated from lands west of the Bosphorus Straits. This not complete population change, but the arrival of new peoples with socio-economic strategies that were highly suited to an environment in which they could prosper and slowly become dominant on the one hand and merging over time with much of existing population on the other; in short, acculturation. From Gordion there is archaeological evidence that can be cited in support of new elements in the EIA population. The last LBA level (YHSS 8) was abandoned with a number of typical late Hittite pottery vessels left in situ, leaving no doubt as to its place in the Hittite empire.28 With regard to the absolute date, it is now suggested that the end of the Hittite culture may have come a generation or so after the desertion of Ḫattuša, if that is to be put at c. 1180 BCE. However, the evidence for this lag is not entirely robust. Not only is radiocarbon dating and calibration subject to statistical error, and it should not be forgotten that the excavated sounding into the LBA level measured only 9 by 14m. While it might be considered likely that this evidence recovered at Gordion is representative of the entire LBA settlement and the transition into the EIA, this cannot be certain. Whatever the exact date, the archaeological evidence clearly demonstrates that the interval between the filling of the LBA building and the first EIA occupation was brief and that the EIA occupation was of a very different nature. There seems no reason to repeat here the detailed evidence for that transition, or for the conclusion that the EIA settlement included a mixture of the local population and newcomers. The newcomers brought with them a ceramic tradition that ultimately, if not directly, came from the Thraco-Balkan region far to the west. Some of the hand-made pottery, as Robert Henrickson has demonstrated, is directly ancestral to Early Phrygian wheel-made types.29 Thus it may not be incorrect to call the newcomers ‘Proto Phrygian’. Archaeological evidence does not permit us to gauge what percentage of the EIA population comprised newcomers, nor to know the extent to which the material culture of the EIA levels at Gordion can be ascribed to them since, as will be discussed later, it is with the Early Phrygian level in the tenth century that distinctive Phrygian architectural and other cultural traits suddenly become dominant. At Gordion, then, it is possible to recognise incoming Early Iron Age immigrants from the West, ceramic production that owes nothing to the earlier Hittite tradition, and a mixed population. It is not, perhaps, possible to go further on archaeological evidence alone. However, the new TKH 1 hieroglyphic inscription, together with resultant new readings and interpretations of the Kızıldağ inscriptions, may offer new insights. The central issue is whether the names in the inscription, Kartapu and Hartapu, are variant spellings of the same king or refer to two kings who ruled several centuries apart. The beginning of TKH 1 is carved in relief while the greater part of the text is incised. The first portion, in relief, names Kartapu son of Mursili while in line 2, incised, we have Hartapu. It is agreed that K/Hartapu are variants of the same name, but there is vigorous debate about whether these different spellings refer to one and the same Great King or if the K and H intentionally signal two kings bearing the same (dynastic) name.30 There is agreement that TKH 1 was carved in the eighth century, very probably in the first half, at a time before Midas had ascended the throne at Gordion, because Hartapu, unlike Mita (Midas), is not mentioned in any Neo-Assyrian inscription.31 It is agreed that the text of the first line of text bears very close comparison with the KIZILDAĞ 4 (KzD 4) inscription which, as Hawkins and Weeden are the most recent to argue,32 has to be dated to the EIA on palaeographic grounds. The first editors of TKH 1 and the directors of the TKH project reject arguments for the early date, favouring instead a single Great King named K/Hartapu who ruled in the eighth century, and conclude that the events referred to in 28 This summary is based on Kealhofer et al. (2019), which provides the most recent overview of the LBA/EIA transition at together with the most recent absolute dates and Voigt/Kealhofer (in press) that gives details of excavated sequence and its interpretation. 29 Henrickson 1994. 30 For detailed discussion see Massa/Osborne (2022) with references. For multiple Neo-Hittite rulers bearing the same name see, for example, Peker (2022), where three rulers of Gurgum are named Halparuntiya, three Larama and two or three Muwattalli. 31 This argument is not watertight, it could also be that the Assyrians had little knowledge of the the Konya Plain. 32 Hawkins/Weeden 2021; Massa/Osborne 2022 both with references. I am most grateful to David Hawkins, Michele Massa, James Osborne and Mark Weeden for providing drafts of their papers and for much discussion that, strong disagreements notwithstanding, have been most friendly. 112 Altorientalische Forschungen 2023; 50(1) TKH 1 and all of the other inscriptions that name Hartapu were more or less contemporaneous.33 The stance taken here is that the arguments adduced by Hawkins and Weeden are the more convincing, not least because it may be thought that both the contrast between relief and incised carving and the variant spellings were intentional devices differentiating between two kings.34 Hawkins and Weeden offer this translation of the first line of TKH 1: ‘Great King Kartapu, Hero, son of Mursili, (is the one) who conquered the land Muska’ As adduced by Hawkins and Weeden, this Mursili was unlikely to have been the LBA Great King Mursili III because his regnal dates are too early. Rather, he was a later, EIA, king of the same name, whether a descendant or a claimant. If this EIA dating is correct, on TKH 1 and KzD 4, we have the first known references to the Land of Muška, a reading not now contested. This immediately brings to mind the eighth century Neo-Assyrian references to Mitā of Muški that is almost universally accepted as equating with Midas of Phrygia. Petra Goedegebuure et al. go further in arguing that the Land of Muška in TKH 1 is to be translated Phrygia. Hawkins and Weeden, on the other hand, see this Land of Muška as a reference to seminomadic tribes of Muški who, since they are not known from Hittite texts, would have been twelfth century arrivals on the western part of the Central Anatolian Plateau.35 Here it is suggested that the Land of Muška defeated by Great King Kartapu in TKH 1 and Hartapu on KzD 4 refers to the territory directly to the north of his kingdom. It is also likely that K/Hartapu not so much conquered the land in the Sakarya-Porsuk basin as stemmed a significant incursion from that direction. We will return to the issue of the territory ruled by Kartapu, and later by Hartapu; suffice it here to suggest that the Neo-Hittite kings with a capital at Türkmen-Karahöyük are unlikely to have been sufficiently powerful to rule over a land that stretched so far to the north as the Sakarya/Porsuk basin. It is my contention here that this land of Muška could very well be equated with Gordion and its surrounding lands, and that these Muški, who were only one of several related groups bearing that name to have migrated into various regions of Anatolia, were the newly arrived portion of the population that have been identified in the EIA levels at Gordion where they had rapidly become dominant.36 If this reconstruction has credence, there are implications for the proportion of newcomers to locals as well as the number of immigrants which, in this case, would have been significant. In other words, the Land of Muška was a threat to the Neo-Hittite state of K/Hartapu that required dealing with by force, and that the victory was of enough importance not only to be inscribed on KzD 4 but repeated several hundred years later on TKH 1 (Table 2).37 If, however, the interpretation of a single eighth century Great King Hartapu son of Mursili turns out to be correct, translation of the Land of Muška as Phrygia would match the Neo-Assyrian evidence from the time of Midas.38 While that solution would remove the name Muški from discussion of the LBA/EIA transition, it would not impact on the general thrust of this paper. Three, possibly five, of the total of eight known inscriptions have K/Hartapu I son of Mursili and four have Hartapu II. The dates for Mursili III, otherwise called Urḫi-Teššub, seem to be too early for the father of K/Hartapu.39 On epigraphic grounds (Hawkins pers. comm.) KzD 4, KrD 1 and KrD 2 cannot be much later than 33 Goedegebuure et al. 2020; Massa/Osborne 2022. 34 TKH 1 is very odd and unexpected. The first line is taller than the rest and the incised signs at left are more widely spaced than those those on the narrower lines below. This could be because the intention was to carve the entire first line in relief. See also the comments in Massa and Osborne (2022). 35 Hawkins and Weeden do not make any connection between the Land of Muška in TKH 1 and the Muški known from Assyrian texts to have been in the southeast of Turkey at the cusp of the Iron Age. Furthermore, they are reluctant to accept the translation Phrygia for what they consider to be an EIA text on KzD 4 (repeated much later on TKH 1). 36 Kopanias (2015) perceptively discussed Muški and associated them with the Land of Maša in the Hartapu inscription. Maša, as a result of TKH 1, is now read Muška, defeated by Kartapu. Wittke (2004) has a rather different perspective, particularly regarding a movement from east to west, which has not found much favour. 37 I am broadly in agreement with the views of Hawkins and Weeden, not least because I agree that Kartapu and Hartapu on TKH 1 are two different people and also because it is likely the unfinished relief carving of line 1 was intended to be an explicit reference to the earlier text. 38 Goedegebuure et al. 2020; Massa/Osborne 2022. 39 Hawkins/Weeden 2021: 396. It is tempting to coin Mursili IV for this Great King, but Mark Weeden points out to me that this would imply that he was a direct descendent of LBA dynasty that ruled from Ḫattuša. While that is not impossible it is very far from certain. 113 Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom the 12th century. How robust is that epigraphic evidence? There are no 11th or 10th century inscriptions from “Cappadocia” with which to make comparisons. Any lowering the date does not, however, resolve the question of where the hieroglyphic tradition survived from the Empire period into the Middle Iron Age (although TKH has now become a good candidate), nor does narrowing the gap between Kartapu and Hartupu from 400 to 300 years resolve the basic problem. Since Muški are attested in the southeast of Turkey in the 12th century, mention of the Land of Muška (presumably located in the Sakarya/Porsuk basin) in TKH 1 and KzD 4 is not to be ruled out on chronological grounds. Problematic is BURUNKAYA (BK).40 The northeasterly location of BK is a better fit with historical reconstructions of the time of Hartapu II referred to in both TKH I and TOPADA but would not sit comfortably with a twelfth century date. While there is no indication in the inscriptions as to the extent of the fighting or the size of the forces on either side, one result would presumably have been the establishment of a border because the concept of the Land of Muška implies the idea of established territory with boundaries. It is not impossible that the northern or northwestern border of the Lower Land in the Hittite Empire Period was reestablished.41 Table 2: Tabulation of occurrences of K/Hartapu I son of Mursili and Hartapu II. According to Hawkins and Weeden epigraphic evidence suggests that KzD 4, KrD 1 and KrD 242 were cut in the Early Iron Age (12th century), the rest in the Middle Iron Age (8th century, before Midas). THK 1 line 1, KzD 3 and KzD 5 are late renderings of the EIA text.43 THK = Türkmen-Karahöyük, KzD = Kızıldağ; KrD = Karadağ, BK = Burunkaya TKH 1 line1 KzD 3 KzD 4 KrD 1 KrD 2 KzD 5 X X X X X ? Great King X X X X X Hero X Mursili son of X son of X son of X ? son of X ? of X Great King X X X X X X X X X K/Hartapu I Hartapu II Hero Muška X TKH 1 line 2 KzD 1 KzD 2 BK X X X X X X X X X The Kingdom of Phrygia Phrygia is a name used only by Greek and later writers. We do not know what the people that we call Phrygians called themselves; neither do we know what name they gave to their capital which the Greeks called Gordion. TKH 1 now makes Muški for the people and Muška for the land distinct possibilities since Muška can now be seen to have been a name used by Luwian speakers as well as by Assyrian ones.44 Furthermore, as pointed out by Hawkins and Weeden, there are no references to this land or people in Hittite texts. Analysis of EIA Gordion, as noted earlier, suggests a mixed population. In the Gordion terminology (Table 1) Phrygian is used first for YHSS 6, now dated as early as the tenth century, when the EIA levels exposed in the limited excavated area on the citadel were swept away and replaced by civic buildings that imply centralised planning by a newly power- 40 If ‘son’ is the correct restoration on BK, either the eighth century text refers to Hartapu I or there were two Hartapus sons of Mursili. Hawkins and Weeden (2021: 391–392) suggested restoration of ‘descendant’ is not entirely satisfactory, see the discussion by Massa and Osborne (2022: 88). 41 On borders see d’Alfonso/Rubinson (2021) and other papers in the same volume. 42 For KrD 2 see Hawkins (2000) CHL I, 437. 43 Hawkins/Weeden (2021: 387–396). 44 I do not see any mileage to be gained in philological analysis of the name Muška. 114 Altorientalische Forschungen 2023; 50(1) ful elite that is also associated with the first megaron type buildings and the earliest tumulus burials.45 Thus the beginning of ‘Phrygia’ may be associated with YHSS 6. This is not to deny continuity in pottery or other aspects of material culture. If the Muški defeated by K/Hartapu were proto-Phrygians we would expect continuity in language and religion, but there is no evidence either way. There are two questions. Firstly, does this abrupt development on the citadel at Gordion represent another wave or waves of migration, and indeed one that might be more closely related to the various Greek myths about Midas? Secondly, what might the territory of Gordion have been in the later tenth century? The answer to the first is very possibly yes, but there is no need here to enter into discussion of the historicity of later Greeks myths. Regarding the second, with two notable exceptions there has been a strong tendency to view MIA Gordion as the capital of a wealthy and powerful kingdom that stretched from the Hellespont eastwards to the upper reaches of the Euphrates River.46 For instance, in the opening paragraph of a paper on the formation of the Phrygian state from the perspective of EIA Gordion Voigt and Henrickson could write “historians agree that in the eighth century BCE, central Anatolia was dominated by people who spoke an Indo-European language, Phrygian”.47 This view had long antecedents, one of its most influential proponents being Ekrem Akurgal who in his seminal 1955 Phrygische Kunst included MIA painted pottery of the Alişar IV and related styles in his corpus of Phrygian art. Others, most notably Machteld Mellink, saw Midas ruling over Phrygians and, east of the Kızılırmak (the Red River), Muški whom she associated with the Alişar IV pottery style.48 Although this association of Alişar IV style pottery with Muški has long been discredited, the idea that the entire land within the bend of the Kızılırmak, the heartland of the Hittite Empire, was part of Phrygia has unfortunately stuck all too often; so much so that the current director of the Gordion Project, could state: ‘Midas, Gordion’s most celebrated king (ca. 740–700), during whose reign the Phrygian kingdom reached its largest extent, occupying approximately one-third of Asia Minor’.49 Looking to the south from Gordion, the entire Konya basin and the Lake District are likewise incorporated into eighth century Phrygia. This is partly because Sams thought that the MIA pattern painted pottery from Akurgal’s sounding at Alaattin Tepe in the centre of the city of Konya was Phrygian.50 Additional evidence that has induced scholars to extend the Kingdom of Phrygia so a far to the south are the roof tiles from Düver, which are surely Lydian rather than Phrygian, and the contents of the Elmalı tumuli that certainly have Phrygian connections, but which perhaps speak more nearly of Lydia and Ionia than of Gordion. There is no need to discuss this further since the identification of Türkmen-Karahöyük with the capital of the eighth century Neo-Hittite King Hartapu, with the high probability that the city is to be identified with Parzuta known from the TOPADA inscription,51 leaves no doubt whatsoever that the Konya basin was not part of the kingdom of Phrygia before the time of Midas. What happened after Hartapu, if he was not a contemporary of Midas, is unknown, but see comments on the Cimmerians below. Another category of misleading evidence has been terracotta revetment and roof tiles. In Turkey the first appearance of such tiles was at Lydian Sardis, and they first appear at Gordion after the Lydian takeover. At Gordion the earliest of these tiles as so similar to those at Sardis that Lydian tile-makers must surely have been involved in their production. Related tiles of later date have been found at Pazarlı to the east of the Kızılırmak and, much further north, at Akalan close to Samsun.52 To the south of Phrygia, many such tiles were illegally excavated at Düver in the Lake District. None of these tiles are to be dated earlier than the sixth century, by which time the Kingdom of Phrygia had ceased to exist, and many are in fact later.53 It is therefore inaccurate to characterise these tiles as ‘Phrygian’. 45 Maya Vassileva noted the introduction of megaron type buildings in YHSS 6 in a paper entitled The Westerness of the Phrygian Culture presented at the Phrygia between East and West conference at Pavia on April 9, 2022. Muscarella (1995) equates the first tumuli with the establishment of kingship in a paper written before redating of the Destruction Level. 46 The exceptions are Genz (2009) and van Dongen (2014). 47 Voigt/Henrikson 2000: 17. 48 Mellink 1965; 1991: 639. 49 Rose 2021: 27. See also Rose (2021b: 309) for the unfounded suggestion that the Kerkenes area “was likely under Phrygian control during the eighth and early seventh century”. 50 Sams 1994. 51 Hawkins/Weeden (2021: 394) who kindly tell me that Par-x-ta would be a more accurate transliteration. 52 The Akalan tiles, however, look closer to those from Greece and speak to influences from Pontus rather than Phrygia. 53 For recent studies of Anatolian architectural terracottas with references see Summers (2006); Summerer 2021. Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 115 Yet another line of evidence has been the distribution of Paleo-Phrygian inscriptions.54 Relevant here is the inscription from Kerkenes Dağ, which is securely dated by its archaeological context to the first half of the sixth century, probably not much before the destruction of c. 547. In a recent paper Rostislav Oreshko has ingeniously spotted the similarity between the name of a king Masaur(a)ḫis(s)as on an eighth century Hieroglyphic Luwian inscription from Zeyve Höyük (PORSUK) and the name Masa Urgitos on the Kerkenes inscription.55 According to Oreshko, these must be the same person because the name does not appear in any other Anatolian corpus. However, given the paucity of Phrygian inscriptions the uniqueness of the name hardly constitutes an argument. Because there is no doubt that PORSUK is to be dated to the eighth century, it is claimed by Oreshko that the site at Kerkenes must also date to the eighth century.56 This totally disregards the undeniable archaeological evidence for the founding of Kerkenes no earlier than the late seventh century, and the Paleo-Phrygian inscription K-1 towards the middle of the sixth, not long before the city was burnt. With regard to the linguistic arguments, Mark Weeden kindly informs me: “If Oreshko’s reasoning were to be correct, the form of the genitival adjective he is referring to would be *Masa(-)Urhissis not (what we have) Masa-Urhissas. Compare Kartapus Mursilissis for Kartapu (son of) Mursili. In fact, if it was correct grammar according to what Oreshko wants it would be *Masas *Urhissis. Masa(-)Urhis(s)as could only be the equivalent of Masa Urgitos under the assumption of a linguistic distortion. Not impossible, but also not necessary.”57 More difficult are the Paleo-Phrygian inscriptions on stone from Alaca Höyük and on the nearby peak sanctuary at Kalehisar.58 It is not possible to date any of these on the grounds of paleography or language. However, there is not a shred of evidence for any significant Phrygian presence east of the Kızılırmak before the late seventh century. It is difficult, therefore, to imagine that these inscriptions on stone should be dated earlier than the sixth century. Written evidence of a different kind are the lead strips with Luwian in cursive hieroglyphic script that were found in the vicinity of Kırşehir, both very probably from Yassıhöyük (Kırşehir) now being excavated by the Japanese.59 While these strips are mobile objects, not necessarily written where they were found, they do provide some evidence that the region to the east of the Kızılırmak and south of the Pontic foothills should be considered largely Luwian speaking Neo-Hittite, a conclusion supported but not proven by the distribution of Alişar IV style pattern painted pottery.60 It is uncertain how much if any of this area within the bend of the Kızılırmak was familiar to the Assyrians and whether or not it was called by them Tabal.61 Thus it can be seen that the model of the Kingdom of Phrygia, generally thought of as synonymous with the Kingdom of Midas, is that of a powerful, centralised, despotic kingdom not dissimilar to Urartu in the high massif of eastern Turkey and Armenia, or indeed Assyria.62 Several recently published maps portray eighth century Phrygia extending as far eastwards as the headwaters of the Kızılırmak in the vicinity of Sivas, south to include the Konya plain and the Lake District, and northeast to the Black Sea at Samsun.63 Some modified, reduced, version of this grand picture may conceivably hold true for the time of Midas and may be more confidently applied to the seventh century growth of a Lydian empire under the aggressive expansionist policies of Alyattes and Croesus. However, the tenth and ninth century Phrygian kingdom with its seat at Gordion must have been considerably smaller, on a par with its Neo-Hittite neighbours, with contested borders or frontiers subject to not infrequent change. It may not be incorrect to see Midas at the peak of his power leading a Phry- 54 Obrador-Cursach 2020, with map 1 pp. 615–616. 55 Oreshko 2021: 292–297. 56 I am grateful to Mark Weeden for confirmation of date of the Porsuk inscription. 57 Personal communication by e-mail, 24 April 2022. Bartomeu Obrador-Cursach kindly informs me that Zsolt Simon (forthcoming) has come to very similar conclusions. 58 Obrador-Cursach 2020: 501–503. 59 Akdoğan/Hawkins 2009; 2010; Weeden 2013. 60 D’Alfonso et al. 2022. 61 Simon 2017. 62 Rose (2021) compares the size of great megarons on the citadel at Gordion with palaces at Ninuwa. 63 The exceptions are Genz (2011) and van Dongen (2014). Recent maps showing an over-extended kingdom can be found in Tüfekçi Sivas (2007: 10); Tüfekçi Sivas/Sivas (2012: 13); Rose (2016, fig.1; 2017, fig.1; 2021, fig. 1). 116 Altorientalische Forschungen 2023; 50(1) gian federation as well as an Anatolian alliance against invading Cimmerian hordes or empire-building Assyrians. Vacillations in the relationship between Midas and the Assyrians as recorded in Assyrian texts are perhaps indicative of his weakness and skilled diplomacy rather than the military might at his disposal. The Cimmerian Factor64 It is not possible to recognize Cimmerian material culture in the archaeological record of central Anatolia.65 Nor has it been possible to associate archaeological levels, particularly the Destruction Level at Gordion, with Cimmerian hostilities. Nevertheless, there are poor and difficult historical sources that inform us of significant Cimmerian action. Assyrian texts mention Cimmerians as active in Tabal, an area that in this context is perhaps best identified with southwestern Cappadocia. In 705 Šarru-ukīn (“Sargon II”) of Assyria was killed in battle against Gurdī the Kulummean. This Gurdī is often taken to have been a Cimmerian but, as Mark Weeden points out, this is far from certain.66 While it is unclear where these highly mobile Cimmerians were located, there is a likelihood that they were somewhere to the east of Tabal and south of Urartu. Unverifiable on present evidence is the Greek tradition that the reign of Midas was brought to an end by defeat at Cimmerian hands while, as noted earlier, the account of his committing suicide by drinking bull’s blood is undoubtedly mythical. Accounts by the Byzantine authors Eusebius and Julius Africanus that Gordion itself was sacked by Cimmerians in 696/5 or 676 respectively have not been confirmed by archaeology of the poorly preserved Middle Phrygian levels. About the same time, c.679, Aššur-aḫa-iddina claimed to have defeated Teušpa the Cimmerian in Ḫubušna, i.e. the area around modern Ereğli situated to the east of the Konya plain.67 Thus, on the one hand we have Assyrian forces far from home on the upland Anatolian Plateau able to defeat Cimmerians while, on the other, these same invaders were victorious over the rich and powerful Kingdom of Phrygia, somehow ending the reign of its greatest king, Midas. Selim Adalı characterizes the arrival of these Cimmerians in western Central Anatolia as a migration rather than bands of mounted warriors moving across the country creating waves of destruction. Migration involves families, movable possessions and animals on the hoof, to which may be added booty from conquest that included humans and animals as well as material wealth. Adalı very plausibly envisages Cimmerians under Lygdamis, Assyrian Dugdammē, controlling the Konya region as can be seen from Assyrian texts of Aššur-bāni-apli (Assurbanipal). This would imply that the Neo-Hittite states that had held sway over the Konya Plain and adjacent areas since their emergence from the debris of the Hittite Empire had fallen under Cimmerian control. Furthermore, it would suggest that Phrygia had not expanded to the south of the borders that were established at least as early as the eighth century, perhaps considerably before. In the 650s the Lydian king Gyges was killed as Cimmerians sacked Sardis. Archaeological evidence for this mid-seventh century fall of Sardis has not yet been identified, very probably because excavations have not penetrated these earlier Lydian levels on any scale. While the city of Sardis in the time of Alyattes and Croesus was mightily defended, it is not known how strong its defences were in the time of Gyges. Thereafter Cimmerians disappear from our sights as the aggressively expansive Lydian kings Alyattes and Croesus transform their kingdom into an empire that stretched from the Aegean Sea to the Kızılırmak including, however loosely, the entire area that had formally been Phrygia. 64 65 66 67 I am most grateful to Selim Adalı for discussion. Makhortykh/Kotova 2021. Weeden 2017: 731. Kulummaean ethnicity remains a mystery. Adalı 2019. Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 117 Phrygian Polities Before Midas A speculative map, Fig. 1, is a graphic attempt to show approximate areas of five Phrygian polities on the western portion of the Central Anatolian plateau before the time of King Midas, i.e. in the ninth and earlier eighth centuries BCE. The extent to which Midas may have asserted direct rule over this area or, alternatively, established some kind of federation in response to outside threats is unknown. Apart from a stretch of the Kızılırmak at the eastern limit of Phrygia there is no evidence, archaeological or textual, for boundaries to these areas. There are two main underlying principles. Firstly, it is assumed that each of the polities was of approximately the same size, at around 100 kilometres across and, secondly, that for the most part borders followed upland topography, i.e. ridges and mountain tops. More detailed discussion of some aspects of the geography is given below, but that too is entirely speculative. Also shown on the map are equally speculative territories of four NeoHittite states to the southeast of Phrygia. While these territories are only approximate estimates or guesses, the map does graphically show that MIA Phrygia was not a single polity and also that only the western half of the Central Plateau could rightly be termed Phrygia. Territories were surely more contiguous than the map shows, and there are huge blanks. Fig. 1: Map showing places mentioned in the text with a speculative attempt to indicate the possible territories of Phrygian polities before Midas and three of the Neo-Hittite polities to the southeast (graphic by Michele Massa). As noted above, by the MIA, and probably long before, there were established borders. While we may not now know where they ran, local farmers and herders did because these boundaries defined who exacted tithes or taxes, could demand corvee labour, and who might call men to arms. Of course, these borders were very fluid, with crossing points marked, if at all, only on frequently travelled routes. More mobile elements of the population would normally have moved more or less freely to and across them. On the other hand, borders aid rather than hinder state administration and facilitate negotiation over competing territorial claims. It is likely that in most cases borders ran along ridges and over hills that were in marginal land frequented by herdsmen and 118 Altorientalische Forschungen 2023; 50(1) hunters. Later authorities, most notably Herodotus, clearly thought the eastern border of Phrygia was the Halys River (Kızılırmak). This seems to be correct, supported by the distribution of Phrygian Grey Ware.68 To be clear, I take this border along the river to coincide with the north-flowing reach of the river between the two major crossing points, north at Kırıkkale and south at Köprüköy by Büklükale. Until recent dam construction this stretch of the river flowed through a gorge that was nowhere easy to cross. Around the bend, in the area that was flooded by the Hırfanlı dam before systematic archaeological survey was done, there were wide floodplains on either side of the shifting river-bed that did not constitute an obstacle except when the river was in flood. Along the upper reaches of the Kızılırmak, from Kesikköprü through Avanos to the headwaters near Sivas, there are many fordable places, and the river seems not to have been a formidable barrier at any time.69 Other borders cannot be reconstructed on the basis of either archaeological or textual evidence. In general, in Anatolia routes and roads at all periods tend to follow the topography which, as an accident of geology, trends east-west with few passes either north through the Pontic Mountains or south through the Taurus Mountains. Rivers on the central plateau, with the exception discussed above, were not formidable unless in flood and thus did not form natural borders. The barriers, and thus the borders, more usually ran along ridges and over mountains, as is indicated in the famous Hittite bronze tablet that sets out the borders of Tarḫuntassa. The eighth century TOPADA inscription tells us that the Parzutean king put his frontier ‘on the mountain’.70 The map offered here (Fig. 1) attempts to reconstruct the territories of five Phrygian cities, Dorylaion, Gordion and Ankara together with those of the Neo-Hittite states of Parzuta (perhaps to be placed at TKH), Tuwana, and so forth. Looking now at historical sources, because the Iron Age peoples of Central Anatolia abandoned cuneiform script and writing on clay tablets, the extensive records of the LBA Hittites are not replicated in the first millennium. Thus, with the exception of a modest number of Neo-Hittite inscriptions on stone in Hieroglyphic Luwian and scarce fragments of lead strips in cursive Hieroglyphic Luwian from central Anatolia, as well as a rather pathetic number of Phrygian religious inscriptions on stone and, in Tum. MM at Gordion, graffiti on wax or on wood, there are no texts, only archaeology.71 Contemporaneous written evidence comes mostly from Assyria, with possibly the occasional mention of Phrygians (Muški) in Urartu if these are not the eastern Muški. Western, Greek, sources are all considerably later. With regard to EIA Muška, which was to become part of Phrygia, we have the Luwian Hieroglyphic inscriptions of KzD 4 and now TKH 1. Excavated archaeological evidence regarding MIA Phrygia is very largely restricted to Gordion, with some additional evidence from Şarhöyük at Eskişehir, Hacıtuğrul near Gordion, Midas City in the Phrygian Highlands and the rich tumuli in the vicinity of Ankara. Of these, only Haspels’ excavations at Midas City are fully published. Thus, when it comes to attempting to visualise what a map of early eighth century Phrygia might have looked like we are mostly in the realm of speculation. Nevertheless, the attempt is worth making as part of a much more general approach that considers Phrygia in relation to what is known about Neo-Hittite kingdoms. In the later ninth and early eighth centuries there are few known cities on the western Anatolian plateau that could be considered major urban centres and seats of power, Şarhöyük at Eskişehir, Gordion and Ankara. Later Phrygian cities, such as Pessinus, Afyon and Kelainai have yet to provide good evidence of Middle Iron Age urban settlement.72 Very little is known of MIA Şarhöyük, ancient Dorylaion, but the Eskişehir plain had sufficient well-watered agricultural land fringed by good grazing land to have supported urban centres. At Şarhöyük it has proved very difficult to obtain wide exposures of the deeply buried Phrygian levels in which timber buildings might be clearly revealed. The site at Fındık (Fig. 2) has not been tested by excavation, but neither its size nor the apparent depth of occupation suggest that it attained urban proportions. Little more is known of eighth century Midaeum (Midaeion), located at Karahöyük some 35km east of Dorylaion where there is an im- 68 Summers (1994) since when the conclusions have not been overturned by the results of excavations at Kaman-Kalehöyük (Omura 2011) and Boğazköy (Genz 2007a; 2007b). I am most grateful to Herman Genz, Kimiyoshi Matsumura, and Sachihiro Omura for much fruitful discussion over many years. 69 Weeden 2017: 723. 70 Hawkins/Weeden 2021: 394. 71 None of the more numerous graffiti on pottery sherds contain historical information. 72 Middle Phrygian pottery said to be like that from Gordion is said to have been found at Kelainai but is unpublished, von Kienlin et al. (2013). Midas City has produced very little that looks as early as the eighth century. Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 119 pressive circular mound almost 500m in diameter with an extensive lower city (Fig. 3). Ancient tradition has it that Midaeum was founded by Midas, but occupation goes back to much earlier prehistoric times. It is thus not impossible that before Midas there were two seats of power in the Eskişehir plain, one at the western end and a second in the centre of the eastern half. Fig. 2: Google Earth image of Fındık Tepe. The Phrygian settlement is on a natural flat-topped hill below the village in the centre of the image to the left of the valley. Fig. 3: Google Earth image of Mydaion. The circular mound and the lower town at right are clearly seen. 120 Altorientalische Forschungen 2023; 50(1) Moving east from the Eskişehir plain brings us to Gordion, the capital of Midas and his kingdom of Phrygia, where massive construction works began no later than the tenth century with the huge artificial terrace, after which the Terrace Building was named, and the yet more stupendous replanning of the citadel that is evident both before and after the Destruction Level. Following the fire, the core of the citadel was raised with clean clay by some 4m while the gate was filled with stone that extended outwards in order to increase the elevated area available for construction of the new gate.73 All of this required labour, carts and draft animals, and thus a sufficiently large territory and population from which those resources could be obtained, presumably by a corvee system. While the scale of earth and stone moving is impressive, building works of scale are not unique in Anatolia, as witnessed by the massive stone and earth defences at Ḫattuša in the LBA or the defences MIA Sardis with the Colossal Lydian Structure as well as modulation of the urban landscape.74 If, as is highly probable, the clay layer at Gordion was connected with river works, not only was the citadel raised but the far more extensive urban landscape was significantly modulated. Hacıtuğrul, 22km east of Gordion on the direct road to Ankara (Figs 4 and 5), appears to be a planned, circular, site on a largely natural hilltop. This important site covers around 28 ha, only slightly smaller than the citadel at Gordion before its western side was cut away by the river. Clearly the morphology of the site was altered by work on a scale that perhaps rivals the remodelling of the citadel at Gordion. Members of the Gordion team who visited the site when more could be seen of the excavations than is visible today were of the opinion that substantial Middle Phrygian defences could be seen founded directly on bedrock, an observation which confirms that the site is on a largely natural hill (tepe).75 More recent surface survey and mapping by a Japanese team documents occupation from the EIA, Phrygian, Lydian and Persian periods. There are two associated tumuli of some size, presumably dating to the Middle Phrygian period.76 No lower town has been reported and there are no obvious indications of such on figures 4 and 5. Given the proximity of Hacıtuğrul Tepe to Gordion it is hardly likely that it was an independent polity while the two tumuli, if they are indeed Middle Phrygian and not later, would reflect elite status of their occupants. Yet further to the east, more or less midway along to road between Gordion and the crossing point of the Kızılırmak at Kırıkkale, lies Ankara. Implausibly said by Pausanias to have been founded by Midas, little is known of the settlement that lies below the remains of continuous occupation from at least the MIA until today. Significantly, Ankara is the only other Phrygian city where elite, richly endowed tumuli that date back at least as far as the eighth century have been excavated.77 It is most unfortunate that the excavations and finds have never been fully published.78 It is the contention here that Ankara was the capital of a Phrygian kingdom in its own right. While this idea cannot be proven, it is perhaps unlikely that the kings of Gordion would have permitted such ostentatious displays of wealth as is seen in the Ankara tumuli unless Ankara was independent of them. It is likely that Midas subjugated Ankara and its territory as well as both Midaeion and Dorylaion to the west. Such expansion and consolidation of power could be understood in the face of external aggression from Assyria as well as from Cimmerians. That is not to deny that Midas himself may have been driven by personal ambition. His leadership of coalitions of Neo-Hittite and even Urartian forces against Assyria attests to his military prowess and brilliance as well as his diplomacy. However, he vacillated in his relationship with Assyria, and was, if our sources can be trusted, killed in battle against Cimmerians, suggesting perhaps that his forces had limitations. With Midas’ death Phrygian power seems to have quickly waned, hinting that it was in fact weaker that is often assumed. 73 Mary Voigt kindly informs me that what is now called the Terrace Building Complex is dated to the third or even the fourth quater of the ninth century while construction of fortifications began about 900 and really massive construction around 850 BCE. 74 For modulation of the urban landscape at Ḫattuša see Schachner (2022); for Sardis see Cahill (2010a; 2010b). 75 I am grateful to Mary Voigt for relaying this information. 76 Yamashita et al. 2013. 77 There are many tumuli at Kerkenes, some of them quite large, which are presumed to date to the sixth century, (see below). Exceptional is the Phrygian tumulus near Niğde, see Akkaya (1987; 1991). 78 For a scant overview see Tuna (2007). Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 121 Fig. 4: Google Earth image of Hacıtuğrul Tepe. The footprint of Ankara University trenches can be made out and the two tumuli are visible. Fig. 5: Digital Terrain Model Hacıtuğrul Tepe. The larger of the two tumuli is above the scale, the smaller at upper right. The scarp to the north with farm buildings at the base might represent quarrying of material for modulating Middle Phrygian portions of the site that are partially built on bedrock (graphic made by Michele Massa). 122 Altorientalische Forschungen 2023; 50(1) Phrygia’s Neighbours This section will begin in the north and proceed in an anti-clockwise direction. Almost nothing is known of the mountainous land to the north of Phrygia in the ninth and eight centuries BCE. It is likely that the northern limits of Phrygia as a whole would have followed the ridges running east to west from around Kızılcahamam staying just north of Nallıhan to somewhere in the region of Bilecik. To the west of the plateau lay western Phrygia presumably with a major centre at Dascylium, a name that may not predate the Lydian takeover,79 destined to become the Persian satrapy of Hellespontine Phrygia. Excavations have largely focussed on the Achaemenid levels with the result that almost nothing is known about the Phrygian and Lydian settlement. To the southwest lay Lydia, less powerful and less extensive before the late seventh century than it was to become in the first half of the sixth. The border must have lain somewhere between Uşak and Afyon, possibly along the tops of the high hills with a crossing at Dumlupınar where the last battle in the Turkish War of Independence was fought over five days. I propose then a line westwards along the mountaintops to the south of Afyon before swinging northeastwards, north of both the Lake District and the Konya plain, on the north of the Karacadağ to reach the Kızılırmak flowing northwards. The Lake District is a blank on the eighth century map, but not so the Konya plain. The Konya plain together with southern and central Cappadocia was Neo-Hittite. Michele Massa et al. write about ‘spheres of influence’,80 a term that may not be the most appropriate for Neo-Hittite kings who ruled over Neo-Hittite populations. It could possibly apply to Phrygian kingdoms in which newly arrived westerners ruled over Anatolians, but even in that case it might be better to avoid terminology that belongs to colonial history of the nineteenth and twentieth centuries AD. Much better, in my view, is to think of core areas and more loosely controlled territories. The TKH inscription mentions the defeat of 13 kings while TOPADA, possibly with reference to the same events, has 11.81 Likewise, we know of 28 Kings of Tabal who paid tribute to Assyria, but it is not known if any of these had their seats to the north of the Kızılırmak or as far west as the Konya plain. It is likely that some of these kings were subjects or vassals of Great Kings; there is no evidence. In any case, at some time in the ninth or early eighth century there were multiple Neo-Hittite ‘kings’ each of whom presumably ruled from his own citadel, but whose power was probably subject to constraints. In the eighth century it would appear that the more successful Neo-Hittite rulers, Hartapus, Warpalawa, Wasusarma and others, were increasing the size of their kingdoms by subjugating lesser rivals. Known capitals include Türkmen-Karahöyük (?Parzuta) and Tuwana. Further north is a potential capital at Harmandalı (Nyssa). These kingdoms had fixed borders or frontiers that were subject to change, perhaps frequently, as a result of conflict between polities as recounted in the TOPADA and TKH inscriptions. My purpose here, however, is to argue that none of these Neo-Hittite territories formed part of Phrygia, even under the rule of Midas. One problem that cannot be solved on present evidence is the location of the border between the Konya Plain and Phrygia. TKH 1 and KzD 4 tell us that K/Hartapu defeated the Land of Muška and surely implies that the Kingdom of K/Hartapu directly bordered the land of Muška. The precise meaning is unclear and I have preferred the translation defeated rather than conquered because it is unlikely that any of the Sakarya river basin became Neo-Hittite territory. It is on that basis that the territories are indicated on Fig. 1. The reason for taking the line north of the Karacadağ, and thus north of the Salt Lake, is that there does not appear to be any Middle Phrygian occupation at Çevre Kale, Yaraşlı, where, it has been suggested, a Lydian fortress was established behind the eastern limit of their empire.82 It is possible, but highly uncertain, that the Lydians constructed this fortress before they had completely taken over Phrygia itself. It is awkward that, as I understand it, the extensive excavations at Büklükale have not revealed evidence for an eighth century Phrygian occupation of the stronghold. This suggests to me that the Phrygians were not capable of holding a fort at such a strategic crossing, it being too great a distance from Gordion and Ankara. It also suggests that, as far as Phrygia was concerned, the main crossing point was at Kırıkkale. Thus, while Phrygia might have claimed the west bank of the Kızılırmak as Phrygian territory, it did not have the resources to control the main crossing at the south end of the gorge. 79 80 81 82 Kaan (2010: 250). Massa et al. 2020, especially 67 Fig. 14. I follow here the translation of TKH 1 by Hawkins/Weeden (2021: 393). Özgüner/Summers 2017. Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 123 After Midas, Lydian Gordion and Phrygian Pteria Whatever actually happened to Midas himself, and regardless of the reality of a Cimmerian sacking of Gordion, Phrygia appears to have entered into a decline and perhaps re-fragmentation. When Phrygian culture reemerges in the late seventh and especially in the first half of the sixth century it does so not at the old capital, now the seat of a Lydian governor whose palace on the edge of the citadel is perhaps now being excavated, but in the Phrygian Highlands in the mountainous triangle between the modern cities of Eskişehir, Afyon and Kutahya (Fig. 6). It was presumably at this period, before the Persian takeover of Anatolia, that the cultic site of Midas City (Midas Şehiri) was provided with defences, huge hidden cisterns, some reached by tunnels from within the circuit of defences, and rock-cut cultic installations.83 There are two huge architectural façades at Midas City itself, the Midas Monument, so-called because the name of a later Midas, king and commander, occurs in the dedicatory inscription, and the Unfinished Monument with decorative motifs that confirm its mid sixth century date. There are other façades in the vicinity as well as in the Köhnüş Valley. Here is evidence for a petty kingdom under Lydian hegemony with an exuberance of Phrygio/Lydian architectural style and inscription in Paleo-Phrygian rather than the Lydian of the Empire. Phrygian language was to continue in use, NeoPhrygian being written in Greek script, with a revival in Late Roman times.84 Nevertheless, regardless of its fame in Greek mythology and the spread of the cult of Cybele from Phrygia to Rome, Phrygia slipped away. In Achaemenid times the Achaemenid satraps were probably at Kayseri (Mazaca) and certainly Sardis (Sparda) and Dascylium. In the Late Hellenistic period the artificial creation of Galatia straddled the Kızılırmak, encompassing portions of Phrygia, southern Pontus and northwestern Katpatuka. Roman Phrygia was never a province. Fig. 6: Map showing the possible extent of the Lydian empire and the ‘Phrygian Kingdom of Pteria to the east in first half of the sixth century BCE (graphic made by Michele Massa). 83 Summers 2018b. 84 Obrador-Cursach 2020. 124 Altorientalische Forschungen 2023; 50(1) While the eastern border of Phrygia almost certainly lay along the Kızıkırmak, and all of Phrygia had been at least nominally integrated into the Lydian empire by the time of the Persian invasion, it is only at Gordion, and possibly Hacıtuğrul (although there are no tiles), that we have evidence of a substantial Lydian presence. Of particular note is the complete absence of evidence, archaeological, historical or mythological, that the Lydians took Gordion by force. Recent archaeological work has added to the evidence for considerable prosperity under Lydian rule, with revamped and new buildings provided with ornate terracotta tile roofing. At the same time there is the long-known evidence for a Lydian garrison at the fort in the defensive circuit known as the Küçük Höyük. Of course, it is possible that battles were fought in the plain surrounding Gordion where Phrygian forces were defeated, but it is perhaps more likely that following defeat at the hand of the Cimmerians and the death of Midas, Phrygian forces were in no condition for war. Indeed, they might well have welcomed a protective Lydian presence in the aftermath of turmoil. Such a scenario would go a long way to explaining Lydian acceptance of the strong display of Phrygian culture, amalgamated with Lydian elements such as the depiction of Lydianstyle terracotta roof and revetment tiles, on rockcut architectural façades in the Phrygian Highlands. At Midas City not only were Lydian architectural terracottas depicted on its monumental façades, but a few pieces of such tiles were excavated. Nevertheless, geometric patterning of the façades themselves looks entirely Phrygian as is the language of the inscriptions. On the other hand, the cultural flow was not all one way, as demonstrated by the adoption of the Phrygian mother goddess Matar/Kybele at Sardis. Ankara was perhaps only loosely controlled by the Lydian regime, but it is dangerous here to build much on negative evidence. Looking now to what is often called Eastern Phrygia, i.e. the territory within the bend of the Kızılırmak, it is unnecessary to repeat here the persuasive arguments for the identification of Kerkenes with Pteria, or for the dating of the foundation of Kerkenes/Pteria on a virgin site to around or a little before 600 BCE.85 Perhaps to be more forcefully argued, however, is that none of the Phrygian cultural material that has been excavated at sites to the east of the Kızılırmak need be dated earlier than the late seventh century, i.e. earlier than the foundation of Kerkenes. The most prolific and impressive evidence is to be found at Boğazköy where Kurt Bittel excavated at the Iron Age gate to Büyükkale the famous statue of a Phrygian goddess accompanied by a pair of musicians wearing short pants, possibly representing a sixth century Phrygian version of the competition between Apollo and Marsyas judged here by Matar.86 The fragmentary sculpture in the Cappadocia gate at Kerkenes now puts this unique piece into a broader genre of sixth century Phrygian sculpture.87 Also at Boğazköy, the defences of the Südburg with stone-faced glacis together with the large megaron-like structure and other buildings within the fortifications are highly reminiscent of Kerkenes.88 Likewise, the sudden and dominant appearance of Middle Phrygian plain, often grey, pottery at Boğazköy is dated by Herman Genz to the same period. At Kaman-Kalehöyük also ‘Phrygian grey ware’ suddenly becomes prolific at the end of the seventh century. Thus, however suspect the account of Herodotus may be, his statement that peace between the Lydians and the Medes at the end of a five-year war fixed the border at the Kızılırmak is supported by archaeology. It was this border on the Halys River that, on oracular advice, Croesus famously crossed in order to destroy a mighty empire. The point of the above discussion is to consider the extent of the territory that might have been ruled from Kerkenes, i.e. the late, eastern, kingdom of Pteria that retained and even exaggerated its Phrygian origins and culture. It seems obvious enough that the western border would have been along the Kızılırmak, as perhaps was the southern border. To the west lay the territory claimed by Lydia, and to the southeast perhaps an expanding Cilicia in contested territory. To the north Alaca Höyük and nearby Pazarlı are to be included, and perhaps also Oluz Höyük where evidence of Phrygian culture is emerging from beneath the Persian period levels. But Oluz is rather far to the north and should perhaps be considered a polity in its own right. Going eastwards, there is no evidence, but my sense from seeing the countryside and terrain is that it is unlikely to have extended past Akdağmaden. Preliminary analysis of metals from Kerkenes carried out by Joseph Lehner does not indicate that 85 Three references will suffice: Summers 2013; 2018a; Summers/Summers 2010. 86 This interpretation of the musicians is strengthened by a similar scene on one of the slabs at Karatepe near Adana, for which see Özyar (2021). For an eighth century date for this relief see, most recently, Novák/Fuchs (2021). 87 Summers 2021. 88 I owe this observation to Andreas Schachner. Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 125 the Akdağmaden mines were a source of ores in the MIA.89 We can perceive, then, a territory that would support the huge new city on the Kerkenes Dağ, provide corvee labour to build its 7km of defences as well as provisioning it with sustenance. Conclusions Firstly, Phrygia before Midas should not be envisaged as a single unified state but, rather as a number of smaller kingdoms that shared many Phrygian cultural traits, of which language and cult were perhaps the most significant. Under Midas, in the eighth century, Gordion grew into an ostentatious capital of an extensive kingdom, becoming an influential player on an international stage. However, its swift demise after suffering a defeat at Cimmerian hands and its subsequent subjugation to Lydia suggest that Phrygia was never as powerful as is frequently thought. The kingdom of Midas did not extend east of the Kızılırmak, ended somewhere north of the Konya Plain and never reached over the Pontic Mountains to the north. Rich and powerful as it was, finds of gold have been rare and the exquisiteness of the carpentry and woodworking is not matched in a wealth of precious metals or copious ivory. Most impressive is the power to command labour that moved vast quantities of clay and stone, and the sophistication of its monumental wooden architecture. Secondly, in Eastern Phrygia, east of the Kızılırmak, there was a new foundation at Kerkenes, almost certainly ancient Pteria, along with smaller but significant settlements at Boğazköy and Alaca Höyük. Founded after the death of Midas and in existence for less than a century Kerkenes may be regarded as a Phrygian swansong before it was crushed between the two superpowers of the day, Lydia and Persia. Acknowledgements: I am extremely grateful to Michele Massa for much discussion and also for making the two maps in GIS according to my speculations. I am also grateful to Selim Adalı, David Hawkins, Peter Kuniholm, James Osborne, Annick Payne, Mary Voigt and Mark Weeden for comments that have resulted in significant improvements. None of the above are necessarily in agreement with the views expressed which, along with any errors, are the responsibility of the author. References Cited Adalı, S.F. (2019): Barbarians at the Gates: The Cimmerian Occupation of Ereğli (Cybistra) and Its Aftermath. In: C. Maner (ed.), Crossroads: Konya Plain from Prehistory to Byzantine Period (9th International ANAMED Annual Symposium), Istanbul, 161–178. Akdoğan, R./J.D. Hawkins (2009): Kırşehir-Yassıhöyük’ten ele geçen luvi hiyeroglif yazılı kurşun levha, Anadolu Mediniyetleri Müzesi’nin Yıllığı 2008–2009, 7–14. Akdoğan, R./J.D. Hawkins (2010): The Kırşehir Letter: A New Hieroglyphic Luwian Text on a Lead Strip. In: A. Süel (ed.), Acts of the VIIth International Congress of Hittitology, Corum 25–31 August 2008, Ankara, 1–16. Akkaya M. (1987): Kaynarca tümülüsü Friğ Çağı bronz eserli-Vases de bronze du tumulus de Kaynarca, Eski Eserler ve Müzeler Bülteni 11, 31– 36. Akkaya, M. (1991): Phrygian Objects from the Tumulus of Kaynarca. In: Br. Le Guen-Pollet/O. Pelon (ed.), Southern Cappadocia Until the End of Roman Times – State of Research. Proceedings of the Istanbul Conference, French Institute for Anatolian Studies in Istanbul, April 13–14, 1987, Paris, 25–27. Akrugal, E. (1955): Phrygische Kunst, Ankara. Berndt-Ersöz, S. (2008): The Chronology and Historical Context of Midas, Historia 57, 1–37. Berndt Ersöz, S. (2019): To Drink Bull’s Blood: An Analysis of the Story of Midas. In: G.R. Tsetskhladze (ed.), Phrygia in Antiquity: From the Bronze Age to the Byzantine Period: Proceedings of an International Conference ‘The Phrygian Lands over Time: From Prehistory to the Middle of the 1st Millennium AD’, Held at Anadolu University, Eskişehir, Turkey, 2nd–8th November, 2015, Colloquia Antiqua 24, 93– 108. Brixhe, C./G.D. Summers (2006): Les inscriptions phrygiennes de Kerkenes Dağ (Anatolie Central), Kadmos 45, 93–135. Cahill, N.D. (2010a): Sardeis Şehri – The City of Sardis. In: N.D. Cahill (ed.), Lidyalılar ve Dünyaları – The Lydians and Their World, Istanbul, 75–105. 89 Joseph Lehner pers. comm. 126 Altorientalische Forschungen 2023; 50(1) Cahill, N.D. (2010b): Sardeis’te Pers Tahribi – The Persian Sack of Sardis. In: N.D. Cahill (ed.), Lidyalılar ve Dünyaları – The Lydians and Their World, Istanbul, 339–361. Carpenter, R. (1966): Discontinuity in Greek Civilization (The J.H. Gray Lectures for 1965), Cambridge. Christiansen-Weniger, F./O. Tosun (1939): Die Trockenlandwirtschaft im Sprichwort des anatolischen Bauern, Ankara. Cline, E. (2021): 1177 B.C.: The Year Civilization Collapsed: Revised and Updated (Turning Points in Ancient History 6), Princeton. d’Alfonso, L. (2019): War in Anatolia in the Post-Hittite Period: The Anatolian Hieroglyphic Inscription of Topada Revised, JCS 71, 133–152. d’Alfonso, L. (2020): Reorganization vs. Resilience in Early Iron Age Monumental Art of Central Anatolia. In: M. Cammarosano et al. (ed.), talugaeš witteš. Ancient Near Eastern Studies Presented to Stefano de Martino on the Occasion of his 65th Birthday (Kasion 2), Münster, 81–101. d’Alfonso, L. et al. (2022): Regional Exchange and Exclusive Elite Rituals in Iron Age Central Anatolia: Dating, Function and Circulation of the “Alişar-IV Ware”, AnSt. 72, 37–87. D’Alfonso, L./K. Rubinson (2021): Borders and Archaeology. In: L. d’Alfonso/K. Rubinson (ed.), Borders in Archaeology: Anatolia and the South Caucasus ca. 3500 – 500 BCE (ANES Suppl. 58), Leuven, 3–35. DeVries, K. (1990): The Gordion Excavation Seasons of 1969–1973 and Subsequent Research, AJA 94, 371–406. Gabriel, A. (1952): Phrygie: exploration archéologique. Vol. 2: La cité de Midas: topographie. Le site et les fouilles, Paris. Gabriel, A. (1965): Phrygie: exploration archéologique. Vol. 4: La cité de Midas: architecture, Paris. Gassner, G./F. Christiansen-Weniger (1942): Nova Acta Leopoldina, Halle. Genz, H. (2007a): Kızılırmak Bölgesi’nde Demir Çağı – The Iron Age in the Kızılırmak Region. In: H. Sivas/T. Tüfekçi Sivas (ed.), Friglerin Gizemli Uygarlığı – The Mysterious Civilization of the Phrygians, Istanbul, 127–140. Genz, H. (2007b): Late Iron Age Occupation on the Northwest Slope at Boğazköy. In: A. Cilingiroğlu/A. Sagona (ed.), The Proceedings of the Sixth Anatolian Iron Ages Symposium Held at Eskişehir, Turkey, 16–20 August 2004 (Anatolian Iron Ages 6), Leuven, 135–151. Genz, H. (2009): Thoughts on the Use of Seals in Iron Age Anatolia. In: H. Sağlamtimur et al. (ed.), Studies in Honour of Altan Çilingiroğlu: A Life Dedicated to Urartu on the Shores of the Upper Sea, 309–317. Istanbul. Goedegebuure, P. et al. (2020): TÜRKMEN-KARAHÖYÜK 1: A New Hieroglyphic Luwian Inscription from Great King Hartapu, Son of Mursili, Conqueror of Phrygia, AnSt. 70, 29–43. Gürsan-Salzmann, A. (2005): Ethnographic Lessons for Past Agro-Pastoral Systems in the Sakarya-Porsuk Valleys. In: L. Kealhofer (ed.), The Archaeology of Midas and the Phrygians: Recent Work at Gordion, Philadelphia, 172–189. Haspels, C.H.E. (1951): Phrygie: exploration archéologique. Vol. 3: La cité de Midas: céramique et trouvailles diverses, Paris. Haspels, C.H.E. (1971): The Highlands of Phrygia: Sites and Monuments, Princeton. Hawkins, J.D. (2000): Corpus of Hieroglyphic Luwian Inscriptions. Vol. 1: Inscriptions of the Iron Age, Berlin. Hawkins, J.D. (2002): Anatolia: The End of the Hittite Empire and After. In: E.A. Braun-Holzinger/H. Matthäus (ed.), Die nahöstlichen Kulturen und Griechenland an der Wende vom 2. zum 1. Jahrtausend v. Chr. (Kolloquium des Sonderforschungsbereiches 295), Mainz, 143–151. Hawkins, J.D./M. Weeden (2021): The New Inscription from Türkmenkarahöyük and its Historical Context, AoF 48, 384–399. Henrickson, R.C. (1994): Continuity and Discontinuity in the Ceramic Tradition at Gordion During the Iron Age. In: D. French/A. Çilingiroğlu (ed.), Proceedings of the Third International Anatolian Iron Ages Symposium in Van, Turkey 6–12 August 1990 (Anatolian Iron Ages 3), London, 95–129. Kaan, I. (2010): A New Discovery in Dascylium: The Persian Destruction Layer. In: P. Mattiae et al. (ed.), Proceedings of the 6th International Congress on the Archaeology of the Ancient Near East. Vol. 2: Excavations, Surveys and Restorations: Reports on Recent Field Archaeology in the Near East (ICAANE 6.2), Wiesbaden, 249–263. Kealhofer, L. (2005): The Gordion Regional Survey. In: L. Kealhofer (ed.), The Archaeology of Midas and the Phrygians: Recent Work at Gordion, Philadelphia, 137–148. Kealhofer, L. et al. (2019): Dating Gordion: The Timing and Tempo of Late Bronze and Early Iron Age Political Transformation, Radiocarbon 61.2, 495–514. Kealhofer, L. et al. (in press): Ancient Anatolia, Cambridge. Knapp, A.B. (2021): Migration Myths and the End of the Bronze Age in the Eastern Mediterranean, Cambridge. Knapp, A.B./S. Manning (2016): Crisis in Context: The End of the Late Bronze Age in the Eastern Mediterranean, AJA 120, 99–149. Kopanias, K. (2015): The Mushki/Phrygian Problem from the Near Eastern Point of View. In: N.Chr. Stampolidis et al. (ed.), NOSTOI: Indigenous Culture, Migration and Integration in the Aegean Islands and Western Anatolia During the Late Bronze and Early Iron Age, Istanbul, 211–226. Kuniholm, P.I. (1990): Archaeological Evidence and Non-evidence for Climatic Change. In: S.J. Runcorn/J.-C. Pecker (ed.), The Earth’s Climate and Variability of the Sun Over Recent Millennia: Geophysical, Astronomical and Archaeological Aspects. Proceedings of a Royal Society and Académie des Sciences Discussion Meeting held on 15 and 16 February 1989 (Philosophical Transaction of the Royal Society), London, 645–655. Kuzucuoğlu, C. (2015): The Rise and Fall of the Hittite Empire in Central Anatolia: How, When, Where Did Climate Intervene? In: D. Beyer et al. (ed.), La Cappadoce méridionale de la préhistoire à la période byzantine, Istanbul, 8–9 Novembre 2012, Istanbul, 17–41. Mahortyh, C.B./N.S. Kotova (2021): “Kimmero-Skifskie” Drevnosti iz Central’noj Anatolii, Vostok Afro-Aziatskie obshchestva istoriia i sovremennost 1, 54–77 (Мakhortykh S.V./N.S. Кotova (2021) “Cimmero-Scythian” Antiquities from Central Anatolia). Manuelli, F. et al. (2021): The Beginning of the Iron Age at Arslantepe: A 14C Perspective, Radiocarbon 63.3, 885–903. Marsh, B. (1999): Alluvial Burial of Gordion, an Iron Age City in Anatolia, JFA 26, 163–175. Geoffrey D. Summers – Resizing Phrygia: Migration, State and Kingdom 127 Massa, M. et al. (2020): A Landscape-oriented Approach to Urbanisation and Early State Formation on the Konya and Karaman Plains, Turkey, AnSt. 70, 45–75. Massa, M./J. Osborne (2022): On the Identity of Hartapu: Textual, Historical and Archaeological Analysis of an Anatolian Iron Age Ruler, AoF 49, 85–103. Mellink, M.J. (1965): Mita, Muški, and Phrygians, AnaAra. 2, 317–325. Mellink, M.J. (1991): The Native Kingdoms of Anatolia. In: J. Boardman et al. (ed.), The Cambridge Ancient History. Vol. 3 Part 2: The Assyrian and Babylonian Empires and Other States of the Near East, from the Eighth to the Sixth Centuries B.C. Second ed., Cambridge, 619–665. Müller-Karpe, A. (2009): The Rise and Fall of the Hittite Empire in the Light of Dendrochronological Research. In: S.W. Manning/M.J. Bruce (ed.), Tree-Rings, Kings, and Old World Archaeology and Environment: Papers Presented in Honor of Peter Ian Kuniholm, Oxford, 253–262. Muscarella, O.W. (1995): The Iron Age Background to the Formation of the Phrygian State, BASOR 299/300, 91–103. Naumann, C. (1893): Vom goldenen Horn zu den Quellen des Euphrat: Reisebriefe, Tagebuchblätter und Studien über die Asiatische Türkei und die Anatolische Bahn, Munich. Novák, M./A. Fuchs (2021): Azatiwada, Awariku from the “House of Mopsos”, and Assyria. In: A. Payne et al. (ed.), Beyond all Boundaries: Anatolia in the First Millennium BC (OBO 295), Leuven, 397–466. Obrador-Cursach, B. (2019): Γορδίας and Γόρδιον: A Remark on Phrygian Onomastics and Ethnics, JIES 47, 545–551. Obrador-Cursach, B. (2020): The Phrygian Language (HdOr. 139), Leiden. Omura, S. (2011): Kaman-Kalehöyük Excavations in Central Anatolia. In: R. Steadman/G. McMahon (ed.), The Oxford Handbook of Ancient Anatolia (10,000–323 BCE), Oxford, 1095–1109. Oreshko, R. (2021): Phrygians in Disguise: Onomastic Evidence for Phrygian-Anatolian Ethnocultural Contact in Hieroglyphic-Luwian Inscription PORSUK and Elsewhere, Res Antiquae 18, 283–316. Osborne, J.F. (2020): The Syro-Anatolian City-States: An Iron Age Culture, Oxford. Osborne, J.F. et al. (2020): The City of Hartapu: Results of the Türkmen-Karahöyük Intensive Survey Project, AnSt. 70, 1–27. Özgüner, N.P./G.D. Summers (2017): The Çevre Kale Fortress and the Outer Enclosure on the Karacadağ at Yaraşlı, AnAnt. 25, 1–16. Özyar, A. (2021): Signs Beyond Boundaries: The Visual World of Azatiwataya. In: A. Payne et al. (ed.) Beyond all Boundaries: Anatolia in the First Millennium BC (OBO 295), Leuven, 476–516. Peker, H. (2022): A Bull Statue with Hieroglyphic Luwian Inscription (MARAŞ 16) in Kahramanmaraş Museum and the Chronology of the Late Hittite Kingdom of Gurgum, Gephyra 24, 1–10. Rose, C.B. (2016): Gordion and the Penn Museum. In: C.B. Rose/G. Darbyshire (ed.) The Golden Age of King Midas. Exhibition Catalogue, Pennsylvania, 3–16. Rose, C.B. (2017): Fieldwork at Phrygian Gordion, 2013–2015, AJA 121.1, 135–178. Rose, C.B. (2021): Midas, Matar, and Homer at Gordion and Midas City, Hesperia 90, 27–78. Rose, C.B. (2021b): Troy and Phrygia During the Iron Age. In: L. d’Alfonso/K. Rubinson (ed.), Borders in Archaeology: Anatolia and the South Caucasus ca. 3500–500 BCE (ANES Suppl. 58), Leuven, 285–320. Rose, C.B. (ed.) (2012): The Archaeology of Phrygian Gordion, Royal City of Midas (Gordion Special Studies 7), Philadelphia. Rose, C.B./G. Darbyshire (ed.) (2012): The New Chronology of Iron Age Gordion (Gordion Special Studies 6), Philadelphia. Rose, C.B./G. Darbyshire (ed.) (2016): The Golden Age of King Midas. Exhibition Catalogue, Philadelphia. Sams, G.K. (1994): The Early Phrygian Pottery (The Gordion Excavations 1950–1973: Final Reports 4), Philadelphia. Schachner, A. (2022): Building for King and Country: Architecture as a Symbol of the Hittite Empire. In: S. de Martino (ed.), Handbook of Hittite Empire: Power Structures (Empires through the Ages in Global Perspective 1), Berlin, 421–466. Simon, Z. (2017): The Northern Border of Tabal. In: A. Mouton (ed.), Hittitology Today: Studies on Hittite and Neo-Hittite Anatolia in Honor of Emmanuel Laroche’s 100th Birthday, Istanbul, 201–211. Slim, F.G./C. Çakırlar (2022): Pigs and Polities in Iron Age and Roman Anatolia: An Interregional Zooarchaeological Analysis, Quaternary International, https://doi.org/10.1016/j.quaint.2022.05.013. Sotiroff, G. (1969): Slavonic Names in Greek and Roman Antiquities (Onomastica 37), Quebec. Summerer, L. (2021): Hoplites in Pants: Contextualizing the Warrior-Friezes from Pazarlı. In: E. Sökmen/A. Schachner (ed.), Understanding Transformations: Exploring the Black Sea Region and Northern Central Anatolia in Antiquity (Byzas 26), Istanbul, 21–68. Summers, G.D. (1994): Grey Ware and the Eastern Limits of Phrygia. In: A. Çilingiroğlu/D.H. French (ed.), Anatolian Iron Ages 3, London, 241–252. Summers, G.D. (2006): Architectural Terracottas in Greater Phrygia: Problems of Chronology and Distribution. In: B. Avunç (ed.), Hayat Erkanal’a Armağan: Kültürlerin Yansıması. Studies in Honor of Hayat Erkanal: Cultural Reflections, Istanbul, 684–688. Summers, G.D. (2013): East of the Halys: Thoughts on Settlement Patterns and Historical Geography in the Late 2nd Millennium and First Half of the First Millennium BC. In: H. Bru/G. Labarre (ed.), L’Anatolie des peuples, des cités et des cultures. Autour du projet d’Atlas historique et archéologique de l’Asie Mineure antique. Actes du colloque international de Besançon (26–27 novembre 2010), Besançon, 41–51. Summers, G.D. (2017): After the Collapse: Continuities and Discontinuities in the Early Iron Age of Central Anatolia. In: A. Schachner (ed.), Innovation versus Beharrung: Was macht den Unterschied des hethitischen Reichs im Anatolien des 2. Jahrtausends v. Chr.? Internationaler Workshop zu Ehren von Jürgen Seeher, Istanbul, 23–24. Mai 2014 (Byzas 23), Istanbul, 257–274. Summers, G.D. (2018a): Phrygians East of the Red River: Phrygianisation, Migration and Desertion, AnSt. 68, 99–118. 128 Altorientalische Forschungen 2023; 50(1) Summers, G.D. (2018b): Notes on Chronology in the Phrygian Highlands: Cultic Installations, Defenses, and Clamp Cuttings, JNES 77, 67–84. Summers, G.D. (2021): Kerkenes Final Reports 1: Excavations at the Cappadocia Gate (OIP 145), Chicago. Summers, G.D. (2022): Kerkenes Final Reports 2: Excavations at the Palatial Complex (OIP 148), Chicago. Summers, G.D./F. Summers (2010): From Picks to Pixels: Eighty Years of Development in the Tools of Archaeological Exploration and Interpretation, 1927–2007, at Kerkenes Dağ in Central Turkey. In: P. Matthiae et al. (ed.), Proceedings of the 6th International Congress on the Archaeology of the Ancient Near East May, 5th–10th 2008, Sapienza, Università di Roma. Vol. 2: Excavations, Surveys and Restorations: Reports on Recent Field Archaeology in the Near East, Wiesbaden, 669–683. Tüfekçi Sivas, T. (2007): Frigler ve Frig Uygarlığı. Phrygians and the Phrygian Civilization. In: H. Sivas/T. Tüfekçi Sivas (ed.), Friglerin Gizemli Uygarlığı. The Mysterious Civilization of the Phrygians, Istanbul, 115–126. Tüfekçi Sivas, T./H. Sivas (ed.) (2012): Frigler: Midas’in ülkesinde, anıtların gölgesinde. Phrygians in the Land of Midas, in the Shadow of Mountains, Istanbul. Tuna, N. (2007): ODTÜ Müzesi’nin Ankara Frig Nekropolü’nde Araştırma ve Kurtarma. In: H. Sivas/T. Sivas (ed.), Friglerin Gizemli Uygarlığı. The Mysterious Civilization of the Phrygians, Istanbul, 99–113. van Dongen, E. (2014): The Extent and Interactions of the Phrygian Kingdom. In: S. Gaspa et al. (ed.), From Source to History: Studies on Ancient Near Eastern Worlds and Beyond. Dedicated to Giovanni Battista Lanfranchi on the Occasion of his 65th Birthday on June 23, 2014 (AOAT 412), Münster, 697–711. Vassileva, M. (2021): Phrygian Onomastic Notes. In: A. Payne et al. (ed.), Beyond all Boundaries: Anatolia in the First Millennium BC (OBO 295), Leuven, 689–701. Voigt, M.M. (2012): The Unfinished Project of the Gordion Early Phrygian Destruction Level. In: C.B. Rose (ed.), The Archaeology of Phrygian Gordion, Royal City of Midas, Philadelphia, 67–99. Voigt, M.M. (2013): Gordion as Citadel and City. In: S. Redford/N. Ergin (ed.), Cities and Citadels in Turkey: From the Iron Age to the Seljuks (ANES Suppl. 40), Leuven, 161–228. Voigt, M.M./R. Henrickson (2000): Formation of the Phrygian State: The Early Iron Age at Gordion, AnSt. 50, 37–54. Voigt, M.M./L. Kealhofer (in press): Observing Change, Measuring Time: Documenting the Late Bronze Age-Iron Age Sequence at Gordion. In: L. d’Alfonso et al. (ed.), Between the Age of Diplomacy and the First Great Empire in Western Asia (1200–900BC), New York. von Kienlin, A. et al. (2013): Kelainai: A Phrygian City between East and West, Collection de l’Institut des Sciences et Techniques de l’Antiquité 1277.2, 221–230. Weeden, M. (2013): A Probable Join to the “Kırşehir Letter”, Anatolian Archaeological Studies 18, 15–17. Weeden, M. (2017): Tabal and the Limits of Assyrian Imperialism. In: Y. Heffron et al. (ed.), At the Dawn of History. Ancient Near Eastern Studies in Honour of J.N. Postgate, Winona Lake, 721–737. Weeden, M./L.Z. Ullmann (ed.) (2017): Hittite Landscape and Geography (HdOr. 1/121), Leiden. Wittke, A.-M. (2004): Mušker und Phryger: Ein Beitrag zur Geschichte Anatoliens vom 12. bis zum 7. Jh. v. Chr. Kommentar zur TAVO-Karte B IV 8 “Östlicher Mittelmeerraum und Mesopotamien um 700 v. Chr.” (TAVO Suppl. B, 99), Wiesbaden. Yamashita, M. et al. (2013): Archaeological Surveys at Hacıtuğrul Höyük in Central Anatolia, Anatolian Archaeological Studies 18, 37–41.