Forging an Empire: The Borders of the Land of Karkemiš According to the Treaty between Šuppiluliuma and Šattiwaza

Abstract

The article analyses the border descriptions of the land of Karkemis according to the treaty between Suppiluliuma and Sattiwaza (CTH 51). We argue that the toponyms that establish the boundaries of Hanigalbat and Karkemis in the treaty were excluvisely on the east bank of the Euphrates; this is contrary to what is usually proposed by scholarship, which locates some of the toponyms on the west bank of the river. Thus, the territory of Karkemis significantly expanded beyond the Euphrates after the establishment of Hanigalbat, while the east bank was subsequently controlled by Suppiluliuma’s son, Piyassili. Pushing the borders beyond the natural limits of the river was groundbreaking, albeit eventually an unfeasible enterprise.

Keywords:

• Karkemis
• Hanigalbat
• Hatti
• Suppiluliuma
• Piyassili
• Sattiwaza
• Treaty
• Borders

Acknowledgments

This research was supported by the Gerda Henkel Foundation (AZ 21/F/19) as part of the project, Forging an Empire: Hittite Imperial Administration from the Mediterranean to the Euphrates. We thank Lorenzo d’Alfonso, Juan A. Belmonte, Gioacchino Falsone and Ran Zadok for their valuable comments and suggestions.

Notes

1 The first attacks of Šuppiluliuma against the Mitanni heartland are probably related in The Deeds of Šuppiluliuma, §26. The conquest of Karkemiš, narrated in §28, is thought to have taken place during the so-called ‘Second Syrian War’ (Bryce 2005: 176–183; Wilhelm 2015: 76).

2 On the duration of the siege, see del Monte 1993: 43; Singer 2011: 741 . Šuppiluliuma’s conquest of the city was evoked later in the Annals of Muršili II, Year 9 (Goetze 1933: 118–119) and in the Treaty between Muwattalli and Talmi-Šarruma, §11 (Beckman 1999: 94).

3 Klengel 1992: 113–114, 121.

4 Irrite is identified with Tell Bandarkhan, 30 km southwest of Harran (Belmonte 2001: 146; Bryce 2009: 292 and 337). Another option is Tell Hajib, 37 km east of the Euphrates (Cancik- Kirschbaum and Hess 2016: 70–71).

5 Following Hawkins (1983: 136), the broken passages have usually been restored thus: ‘And all the cities of [the land of Karkemiš–the cities of Karkemiš], Murmurik … (etc.)’.

6 After Šurun comes the conjunction ù, ‘and’ (l. 17’). Based on the parallel formulation of l. 18’, it was perhaps followed by an additional toponym, which was accidentally omitted by the scribe of the current manuscript. Baṣiru (Tell Bazi) was suggested as an option for the missing toponym (Otto 2009: 172; Sallaberger, Einwag and Otto 2006: 9).

7 The river crossing at Tell Ahmar was still used by the Mosul-Aleppo caravan route until the construction of the modern bridge 10 km downstream at Qara Quzaq (Egea Vivancos 2005: 85; Röllig 1997: 287).

8 There is no reason to consider that Marmaruk(um) was close to Alalaḫ. Even though Casana (2009: 22) states that there were two different towns with similar names, one Marmaruk in Mukiš and another ‘far to the east’, it is known that Yamḫad held lands and cities far and beyond the Euphrates (Lauinger 2015: 162–186). Hence, it can be posited that the Marmaruk(um) of the Alalaḫ sources was probably the Murmuriga of the Treaty between Šuppiluliuma and Šattiwaza.

9 An earlier identification was made by Frans Wiggermann in a lecture in London in 2011.

10 Consider the following locations for Murmurik: Tell Ta>alik, 9 km southeast of Karkemiš and 6 km east of the Euphrates, one of the few sites on the east bank reported to have a Late Bronze stratum (Einwag, Kohlmeyer and Otto 1995: 105). Closer to the riverbank is Shiyukh Tahtani, 10 km south of Karkemiš and 5 km southeast of Shiyukh Fawqani, in a position that better fits the written sources. However, excavations at Shiyukh Tahtani have found little Late Bronze remains (Falsone and Sconzo 2012: 170; Fink 2016: 21, no. 119).

11 It was thought that the Tell Fray tablets were Middle Assyrian, but a recently published tablet from the site by Wilhelm (2018) exhibits remarkable similarities to the Emar Syro-Hittite tablets, ca. 13th–12th century.

12 Marina of ‘Bit-Adini’ and ‘of the land of Ḫatti’ appears in two Neo-Assyrian epigraphs (for the sources, see Lipiński 2000: 175–176, 189–190; Morandi Bonacossi 2000: 376–377, n. 139; Barnett et al. 2008: 31, 45, 111, 139). A third inscription mentioning Marina of Bit-Adini has also been recorded (Barnett et al. 2008: 63,176).

13 Kummuḫ lay 92 km north of Karkemiš as the crow flies, on the west bank of the Euphrates, in the land of Išuwa (Hawkins 1980–1983: 338–340; 1995: 92–94; Forlanini 2004: 415). Kummaḫa, a city in the land of Ḫayaša, is also out of the question (see Röllig 1997: 286; cf. del Monte and Tischler 1978: 220–221; Otten 1980–1983: 334).

14 Röllig (1997: 286–287) suggested locating Kumaḫu either at Birecik (20 km north of Karkemiš) or downstream at or near Til Barsip (see also Alexandrov and Sideltsev 2009: 72).

15 The location of Araziqu at Tell aj-Hajj was proposed because of the association between the Roman fort of Eragiza (allegedly a metathesised Araziqu), named in the Tabula Peutingeriana, with Tell al-Hajj/Abu-Hanaya (see Stucky 1972: 40; K. Miller 1964: 759; Forlanini 2004: 415,

n. 63; Tenu 2009: 204; Miller 2012: 357, n. 21). Araziqu (^uruA-ra-zi-ga) appears in two Syro- Hittite deeds from Emar (Arnaud 1987: no. 13: 13; Beckman 1996: no. 76: 4). Yamada (2011: 201–202) infers that Araziqu was under Hittite control at that point. The Emar Syro-Hittite deeds date approximately to 1275–1175 BCE; it is not possible to date the Emar documents with greater precision.

16 Grayson 1991: 25, vi 63–65: i-na KUR Mi-ta-ni ù i-na ^uruA-ra-zi-qi ša pa-an KUR Ḫa-at-te, ‘in the Land of Mitanni and at the city of Araziqu, which is before the land of Hatti’. The deed was later echoed by Aššur-bēl-kala (Grayson 1991: 103, iv 4–5; see Forlanini 2004: 415, n. 63; Cancik-Kirschbaum 1996: 104; 2009: 141; Röllig 1997: 286).

17 It is not clear if the city is related to the KUR A-ra-zi-qí (or KUR A-ra-zi^ki) recorded just once in the Qaṭna texts (Richter 2007: 307). In Alalaḫ VII, Araziqu (^uruA-ra-zi-iq^ki) is also found in the year formula – ‘The year when Niqmepa conquered Araziq’ (Belmonte 2001: 31; Zeeb 1998: 842). Although it is conceivable that the king of Ḫalab crossed the Euphrates to Araziqu, we are possibly facing homonymous GNs, as suggested by Miller (2012: 357, n. 21), since an Arazik(na) is listed among the GNs near the Orontes River in Thutmose III’s list (Helck 1971: 141, no. 139; Astour 1963: 223, no. 139; Belmonte 2001: 31). This probably is the^uruAr-zi-ga-na recorded in a letter from the king of Karkemiš to Ammištamru II (Nougayrol 1968: no. 27, ll. 42–53 [RS 20.22]; van Soldt 2005: 11, n. 55). The city ^uruA-ra-gi-zi (Malbran-Labat 1991: no. 4; RS 34.131) is not related.

18 Röllig (1997: 287) suggests an identification with Tell Karus, although only Neo-Assyrian levels have been found there (Einwag 2000: 311, 314–315; cf. Bunnens 1999: 609, n. 22; Lipiński 2000: 168–169).

19 The city is also recorded as mdkn in Ugarit (RS 24.285; Belmonte 2001: 196).

20 Pitru appears in Thutmose III’s list (Helck 1971: 146, no. 280; Astour 1963: 231, no. 92; Zadok 1995: 277). The Kurkh Stele (Grayson 1996: 19, col. ii, ll. 35–37) situates Pitru (^uruPi-it-ru) on the Sajur and Mutkinu (^uruMu-ut-ki-i-nu) on the east bank of the Euphrates (Cancik-Kirschbaum and Hess 2016: 14).

21 Note that no Late Bronze levels were recovered at Tell ‘Abr (Tenu 2009: 205; Boese 2009: 74). Other suggestions for Mutkinu are Tell Hamis and Ja'dat al-Mughara, both some 10 km southeast of Til Barsip (see, respectively, Lipiński 2000: 168; Boese 2009: 75). However, they are too far from the Sajur junction. Excavations at Tell Hamis did not reveal Late Bronze levels (Matilla 1999: 219). Tell ‘Abr, Tell Hamis and Ja'dat al-Mughara are currently submerged (Tenu 2009: 205–206).

22 Neo-Assyrian Surunu/Saruna has been identified with Ṣawran, 15 km east of Ḫazazu, modern Azaz (Bagg 2007: 223–224, with reference to previous studies; cf. Yamada 1994: 262, n. 7; Miller 2001: 85–86).

23 An Old Babylonian seal (unprovenanced, but perhaps from Alalaḫ?) mentions Šurunu (^uruŠu- ru-nu^ki), with no further information about this city (Dietrich and Loretz 1969: 214; Eidem 2011: 19).

24 Singer (2008: 719, n. 41) equated Šurun with […]x(-)šu-ru-wa-an-na appearing in a Hittite letter from the Hittite-Assyrian correspondence (Mora and Giorgieri 2004: 103; CTH 187). There is no proof that the sign sequence refers to Šurun, but it is certainly possible. The letter mentions that some cities were given to the king of Karkemiš (Yamada 2011: 207).

25 E.g., in the Euphrates-Ḫabur confluence was Suru, which differs from Šūru (modern Savur), upstream the Ḫabur (Kessler 1980: 57–66; Fales 2012: 105–106).

26 In the Tabula Peutingeriana Sirrin is recorded as Serre (Matilla 1998: 299).

27 Röllig (2009: 268–270) links Ṣirina with Ṣirani (^uruṢi-ra-ni) found in an Arslan Tash inscription.

28 The Sirrin citadel is largely covered by the modern village. Morandi Bonacossi (2000: 376, n.

136 ) lists Sirrin as one of the Late Bronze sites in the area between Karkemiš and the Tishrin Dam (see also Einwag 2000: 314–316; Egea Vivancos 2005: 460).

29 The Treaty of Šuppiluliuma and Šattiwaza, §12, rev. 30’: ù UR^{Udidli.ḫi.a} ša ^mŠat-ti-ú-a-za <a-na> ^mPí-ia-aš-ši-íl-li [it-t]a-ad-di-šu ṣa-bur-ta mi-im-ma la-a ú-ba-‘-a, “And the cities which Šattiwaza has given to Piyaššili shall not plan any evil”.

30 The Treaty of Šuppiluliuma and Šattiwaza, §12, rev. 33’–34’: ù mi-nu-me-e URU^{didli.{ḫi.a}}[ša ^m]Šat-ti-ú-a-za ša i-na a-aḫ ^ídUD.KIB.NUN.NA GAR-nu an-nu-ti-ma li-qí-el-lu-u ù URU-lì ša-na-a i-na a-aḫ [^ídU]D.KIB.NUN.NA la-a iṣ-ṣa-bat, “And whatever cities [of] Šattiwaza which are located on the bank of the Euphrates, he may retain, but he shall not seize any other city on the bank of the Euphrates”.