Navigation – Plan du site

AccueilNumérosXXVIIINew Inscriptions in Aramaic/Early...

New Inscriptions in Aramaic/Early Syriac and Greek from the Cemeteries of Edessa

Bekir Çetin, Müslüm Demir, Alain Desreumaux, John Healey et Peter Liddel
p. 119-141

Texte intégral

The authors have benefited from the comments and suggestions of Simon Brelaud (UMR Orient et Méditerranée) and Professor Polly Low (University of Durham) and thank both these scholars for their help.

Fig. 1: Map of Edessa

Fig. 1: Map of Edessa

J. Healey

1Recent years have seen the uncovering of two major cemeteries in Şanlıurfa/Edessa in two areas of the city (see Fig. 1, Map): the first one, in the Kızılkoyun area to the east of Haleplibahçe Caddesi, which runs north from the citadel area and is adjacent to the recently excavated Roman Baths, the villa where the Haleplibahçe mosaics (dated to the 4th/5th cent. AD) were found and the site of the new Archaeological and Mosaic Museums; and the second one, on the slope of the citadel hill to the west of the citadel itself, an area called Kale Eteği (= “below the citadel, foot of the citadel”).

2It seems that both areas lay originally outside the walls of the Roman-period city and we can assume that such a location was chosen for burials for cultural/religious reasons, the dead being separated from the living and away from temples, etc. It should be noted, however, that the baths and villa found to the west of the Kızılkoyun tombs show that the city expanded, at least later, to the west. Until recently, both Kızılkoyun and Kale Eteği were part of an unplanned urban expansion of poorly built houses. These were expropriated by the local authority and demolished as part of a plan to make Şanlıurfa an attractive centre of tourism and, from 2012, rescue excavations have been carried out by the Şanlıurfa Museum Directorate and the General Directorate of Cultural Heritage and Museums under the chairmanship of Celal Uludağ, Director of the Şanlıurfa Museum, and by the archaeologists Müslüm Demir and Bekir Çetin (see Çetin, 2017; Demir, 2019).

Other cemeteries outside the city are well known, in Şehitlik Mahallesi and Karaköprü (to the north), Vadi Manci (to the west), Yakup Mahallesi and Kırk Mağara (to the southwest) and Eyüp Mahallesi (to the south) (Sinclair, 1990, 15-16, including map).

3The main focus in this article (A below) is on three tombs in the Kızılkoyun necropolis, which have been excavated systematically. They are numbered M 53, M 54 and M 55. Our purpose is to publish and comment on the inscriptions, numbered here 1, 2 and 3. The Kale Eteği excavations also produced a number of inscriptions which are published elsewhere (Uludağ & Önal, 2017): in part B we list these and add supplementary notes on them.

Recently Excavated Tombs in Şanlıurfa and their Inscriptions

The Kızılkoyun (= “red sheep”) necropolis (Figs. 1, 2)

Fig. 2: General view of the Kızılkoyun necropolis

Image 10000000000002A800000158B7B40D1F2D0912DF.tif

J. Healey

  • 1 See also discussion in Luther, 1997, pp. 154-55 and n. 211; Trombley & Watt, 2000, p. 25 n. 122.

The ancient Kızılkoyun necropolis is located in the Yeni Mahalle neighbourhood, which is partly built upon it. To the north is the Bediüzzaman Cemetery and to the west is the Haleplibahçe area where the baths and villa referred to above are located, the villa containing important mosaics (Üney, 2015; Önal, 2017). To the south are the Balıklıgöl (Fish Lake/Pools and public gardens) and the Castle/kale. Traces of the original west gate of Edessa, later disused, have been observed in earlier surveys (see Segal, 1970, p. 186 n. 1, 249). This gate is mentioned in Syriac literary sources as “the gate of the vaults (Syriac kappē /ܟܦ̈ܐ)” (Martyrdom of Habib the Deacon, Cureton, 1864, p. ܦܓ , l. 22, where the gate is connected with the burial-place created by or for Abšlāmā son of Abgar; see also (Pseudo-)Joshua the Stylite, Chabot, 1927, pp. 254 l. 15, 256 l. 8; Chronicon ad annum Christi 1234 pertinens, Chabot, 1920, p. 220 l. 14). It is not clear what the kappē are, but they could be connected with the tombs to which the gate gave access (Segal, 1970, p. 185 n. 1; Duval, 1892, p. 14).1

The Kızılkoyun necropolis is on a steep and rocky limestone outcrop and it is arranged in four terraces, each terrace having burial chambers built into it, in a dense array. This necropolis area was first excavated in 2014; further seasons followed in 2016-17. The first season unearthed 64 rock-cut burial chambers and the second revealed 11 chambers, thus a total of 75. The rock chambers excavated are rectangular in shape and contain arcosolia. Tombs of the following types are found: (a) single-room burial chambers with three arcosolia and klinai (i.e. resting-places for the dead person); (b) multi-room burial chambers with arcosolia and klinai and (c) single-room burial chambers with a dromos, i.e. a corridor or steps leading down to the tomb (for fuller discussion see Demir, 2019, pp. 252-255). The entrances are in many cases simple, but some tombs have decorated façades, with columns, capitals, pediments, etc.

4The most notable of the tombs are those numbered M 53, M 54 and M 55, all three of which have pits in front of them for liquid offerings (Demir, 2019, pp. 255-256).

5Little of this cultural heritage was found undisturbed, since the chambers were used as dwellings in modern times. However, the terracotta mould-made lamps that emerged from the grave chambers can be dated between the 1st and 4th centuries AD. Such dating is confirmed broadly by the epigraphic data. Three inscriptions, one in Edessan Aramaic, one a bilingual in Aramaic and Greek and a third in Greek alone were found in or in front of tombs M 53, M 54 and M 55 (the last of which also has some barely legible Syriac inscriptions on its exterior) (Fig. 3).

Fig. 3: General view of tombs M 53, M 54 and M 55

Fig. 3: General view of tombs M 53, M 54 and M 55

B. Çetin and M. Demir

Syriac/Edessan Aramaic Inscriptions from inside Kızılkoyun Tomb M 53

6The first chamber of M 53 is 170 cm below the surface at the entrance. The tomb has two chambers and it must have been temple-fronted, like M 55, since column bases and fragments were found in front of the entrance. There are two doorways, one in front of the other, at the tomb’s entrance on its west side. The first doorway is made of rectangular limestone blocks, with a lintel. Behind this door there is a second, arched door. In the first chamber, on the north and south walls, there are stone cist-graves inside arcosolia. They are carved out of the bedrock in the same way as the arches above them and each was covered with three rectangular stones (length 118 cm, width 75, 60 and 75 cm, thickness 15 cm). The gaps between the stones which covered the cists (but not the fronts) were plastered.

7The second grave-chamber, entered through an archway, is 20 cm higher in level than the first and accessed by means of a step. On the north, south and east walls of this second chamber there are arcosolium-type burial places. The one on the east side was built to look like a temple façade, with Corinthian columns supporting a pointed (“open pediment”) arch which is decorative rather than structural. Inside the arcosolium there is a cist-tomb carved out of the bedrock with two cover-stones with acroteria on the corners (dimensions of each stone: length 228 cm, width 62 cm, thickness 7 cm). Close to the south arcosolium of this second chamber, there is an east-west oriented sarcophagus (length 220 cm, width 80 cm, height 120 cm) with a cover which has undecorated corner acroteria.

8The north and south arcosolia were arranged in a similar way. The arcosolia were carved out of the rock and then stone coffins were built into the space. These were closed with monolithic covers (length 242 cm, depth 112 cm, thickness 34 cm) to create klinai. The gaps between the stones at the front and on top were plastered. The archaeological evidence suggests the bodies of the dead were placed on these klinai: there were no bone remains in the enclosed spaces below them, though the stones at the front had been broken through at an early date, probably by tomb-robbers.

Fig. 4: General view of the arcosolium with inscription in M 53

Fig. 4: General view of the arcosolium with inscription in M 53

B. Çetin and M. Demir

Fig. 5: The Syriac/Edessan Aramaic inscription in M 53

Fig. 5: The Syriac/Edessan Aramaic inscription in M 53

B. Çetin and M. Demir

9On the stones that close off the lower part of the grave on the south side of this second chamber, there is a four-line inscription written vertically in Syriac (Figs. 4, 5).

Dimensions of the Inscription: Full dimensions of stone bearing inscription, height 48 cm, width 124 cm, thickness 20 cm.
Letter heights: 2-4 cm, extending to 5 cm in letters like ܬ. Space between lines 3-5 cm.
Although the inscription is undamaged, it is crudely scratched into the stone and in its third line it is difficult to read. The letter {ʾ} is particularly irregular. In line 2 it appears to have an additional stroke. The stroke to the right probably belongs correctly to the letter-form: its diagonal stroke reverses to the left at the bottom and this is a feature of every {ʾ} in the inscription. The diagonal stroke to its left is probably accidental or extraneous, perhaps a first attempt at a vertical. We may note also some extraneous supralinear marks in this line, again possibly accidental.
The personal name in line 3 is problematic. The {m} is reasonably clear, but the letter which follows it has a very long upper line and this means it can only tentatively be read as {r} or {d}. There is little doubt about the reading of bytʾ at the end of the line. We are thus left with a probable mrbytʾ (see below).

1. hdʾ ʿrsʾ
2. dzʿwrʾ br
3. mrbytʾ
4. wbnwhy

“This is the resting-place of Zᵉʿōrā son of Mārbaytā and his sons”

10The personal name mrbytʾ would be unique, but it is plausible in Syriac, with the meaning “Lord of the House”. It would be analogous to the common noun-phrase ܪܒ ܒܝܬܐ meaning “steward”.

11The name Zᵉʿōrā is certain and, remarkably, the same name appears also in the bilingual inscriptions from M 54 (below): Greek Ζοώρα or Ζοώρας.

12The word ʿrsʾ appears to be the noun ܥܪܣܐ of classical Syriac, which means “bed”, though it can be used in a funerary context for “bier”. The word is not known in early Syriac epigraphy with this meaning, but it is found in Palmyrene Aramaic in a funerary context: PAT 0053, where ʿrsʾ appears in a monolingual (funerary) inscription dated AD 267. Palmyrene Aramaic also provides us with a bilingual in which ʿrśʾ= κλίνη (in a non-funerary context referring to a couch for a deity): PAT 0258= CIS II 3912, dated AD 134. We may note that in classical Syriac the word is spelled with ܣ, while in Palmyrene spellings with {s} and {ś} are found. On the use of ܣ for /ś/ in early Syriac see Drijvers & Healey, 1999, pp. 22-23.

13On vertical writing of Syriac/Edessan Aramaic see further below.

A Greek/Aramaic bilingual inscription on a broken stone stretcher (probably a lintel) from the doorway of Kızılkoyun Tomb M 54 (Figs. 3, 6)

Fig. 6: Bilingual inscription from outside tomb M 54

Fig. 6: Bilingual inscription from outside tomb M 54

B. Çetin and M. Demir

14Tomb M 54 has a stepped entrance on the west side. Before the excavation, there was fill soil and accumulated stone debris on the steps of the entrance. When the debris was taken away, six steps appeared. As the work continued here, it was understood that the steps had been made in a later period, using reclaimed stones. In the infill on the steps there was found a broken stone block, carved in the shape of a tabula ansata (“tablet with handles”) and bearing a bilingual inscription (length 158 cm, width 46 cm, thickness 23 cm).

15Immediately in front of the entrance there are square column-bases on the left and right, the presence of which indicates that this tomb had a temple-style frontage (see M 53 and M 55). There are two doors at the entrance, one in front of the other, the first made of rectangular limestone blocks and a lintel, and the second a cylindrical (tambour) door. On the front of the lintel of the outer door there is carved a chi-rho symbol or Christogram (☧) encircled by a wreath tied at the bottom.

16The tomb has two rectangular-shaped rooms. In the first, on the north and south walls there are cist-tombs inside arched arcosolia. The entrance to the second chamber is arched and it has three arcosolia, one on each wall, with a triangular pediment above the one on the eastern wall. (For pictures see Demir, 2019.)

17The bilingual inscription has four horizontal lines in Greek, and four vertical lines in the Syriac script. The dimensions of the stone conform to the measurements of the gap left by the lintel above the door: it must have fallen from that position at some stage.

18It is clear that a very small part of the stone is missing (originally between the two parts which were still in situ) and this results in damage to the first line of the Aramaic text. It is reported verbally that this missing fragment of the stone was transported to the nearby Archaeological Museum. At the time of writing, the two main parts of the lintel are located in situ outside the tomb.

  • 2 For the depiction of tabulae ansatae as frames in order “to challenge the two-dimensionality” of th (...)

19The lintel has been well dressed and has the form of a tabula ansata.2 Inside this a bilingual inscription is carefully, even elegantly, carved. That the texts are meant to be a bilingual is clear from their content. It is striking, however, that the Greek is written horizontally, but the Syriac is written vertically (starting at the top of the stone). It is well known that Syriac was commonly written this way in manuscripts and in early inscriptions there are good examples of vertical writing (see Desreumaux, 2009; Briquel Chatonnet, 2019; Healey, forthcoming). In early inscriptions, e.g. at Sumatar Harabesi, the vertical writing may sometimes arise simply because of lack of horizontal space, but even where there is plenty of space, the writing is not always horizontal. Our newly discovered bilingual text is evidence of the combination of horizontal Greek with vertical Aramaic/Syriac writing.

20Despite some initial hesitations, it soon became clear that there were really no serious problems in understanding the Aramaic or Greek text.

21Dimensions: Full dimensions of lintel: width c. 201 cm, height 47 cm, thickness 26 cm. Syriac part: line length of the four lines 28 cm (to break), 37 cm, 36 cm, 43.5 cm. Greek part: height of inscription 25 cm; line length of the four lines: 44 cm, 53 cm, 54, cm, 56.5 cm; space between lines 2-3 cm.
Letter-heights: Syriac standard-height letters 3-5 cm, extending to 7 cm for letters like ܬ. Greek letters 5.5-6.0 cm.
Lettering: the Greek features broken-bar alpha, lunate sigma and epsilon, curved omega.

Aramaic

1. hnʾ byt ʿ[lmʾ]
2. dzʿwrʾ wdbyʾ
3. wdbwʾytws
4. br ʾny wdbnwhy
5. ☩

“This is the house of [eternity?] of Zᵉʿōrā and of Bīyā and of Boʾēthos son of Inī and of his children”

22Epigraphically, one remarkable feature is the lack of a dot distinguishing {d} and {r} (as is normal in the early Edessan/Old Syriac inscriptions, though the first Christian Syriac inscriptions known until now already have diacritics to distinguish {d} and {r}: see the Nabgha inscription of 407/8 [Sabbagh et al., 2008] and the Rabbūlā inscription of 425/6 [Briquel-Chatonnet et al., 2008]). Another noteworthy feature is the form of {z} in the second line. The orientation of this letter is unusual and may suggest that the mason was giving close attention to the spacing, since the sloping {z, ܙ} (extending under the upper stroke of the {d, ܖ} and above the right-hand end of the following {ʿ, ܥ}) creates well-judged and elegant spacing throughout the line. Otherwise, where there are initial doubts, they can be easily resolved by reference to the Greek of the bilingual: e.g. the {b} in byt, byʾ, bwʾytws, br and bnwhy, which is angular and could be read as {k}, though the contexts and the Greek equivalents demand that it be read as {b}.

There is one lacuna in the text, in the phrase byt ʿ[lmʾ] in the first line. This lacuna arises from the missing splinter of stone referred to above. As restored, byt ʿlmʾ means literally “house of eternity”. In the Greek a typically Greek expression is used for “tomb”, μνῆμα, which basically means “memorial, monument”, but the Aramaic phrase byt ʿlmʾ is used commonly in Aramaic for “tomb”, both in known Edessan Aramaic/Old Syriac inscriptions (Drijvers & Healey, 1999, As7, As9, Am2, Am3, Am5, Am6, Am7; Önal, Ercan, & Desreumaux, 2013, p. 15) and in later Christian Syriac inscriptions (the Mal Mağarası inscription from near the site of Deir Yakup close to Edessa dated 493/4 [Sachau, 1882, pl. I, pp. 159-160: nr. 4; Segal, 1970, p. 186 n. 4; Güler, 2014, p. 150]). Note byt ʿlmʾ also in the Xantos and Abdu mosaics from Kale Eteği below (unfortunately not dated, but certainly of the 3rd century). The 5th century appears to be the terminus ante quem for the currency of this phrase. It also appears in other branches of Aramaic (e.g. Nabataean: Negev, 1971, pp. 50-52, l. 1; Jewish Aramaic: P. Yadin 7: 15, etc., Yadin et al., 2002, p. 99; see Sokoloff, 2002, p. 217; 1990, p. 95, e.g. Palestinian Talmud Moʿed Qaṭan 1: 80b; Palmyrene: PAT 0548, 0555 and 40 + times). Indeed, it is so common and widespread that one cannot draw any conclusions from it about belief in the afterlife: it is simply the expression used for “tomb”.

23We may note, in passing, the appearance of a Christian cross at the end of the Aramaic (line 5) and the same form of cross appears in line 1 and twice at the end of the Greek. These symbols almost certainly indicate that the tomb owners were Christians, though there is nothing either in the Aramaic or Greek texts which is explicitly Christian. (The personal names are considered separately below.)

Greek

1. ΤΟΥΤΟΤΟ ☩
2. ΜΝΗΜΑΖΟΩΡΑ
3. KΑΙΒΙΑΣΚΑΙΒΟΗΘΟΥ
4. ΙΝΝΕΟΥΚΑΙΤΕΚΝΩΝ
5. ☩ ☩

Τοῦτο τὸ
μνῆμα Ζοώρα
καὶ Βίας καὶ Βοήθου
Ιννεου καὶ τέκνων.

“This is the tombstone of Zoōra and Bia and Boēthos son of Inneos and children.”

24The τοῦτο τὸ μνῆμα + genitive formulation is paralleled in other Greek inscriptions from the East such as SEG 8.15 from Diocaesarea/Sepphoris (Palestine) of the 4th/5th century AD. One striking grammatical feature is the absence of definite article or possessive pronoun with τέκνων in line 4. One might have expected “and their children”: καὶ τῶν τέκνων αὔτων. Parallels (as in the Kale Eteği inscription on plaster, see below) sometimes have a longer expression of this kind (e.g. IGLSyr 5.2030 and 2044). But καὶ τέκνων is attested on an epitaph of the period 4th-7th century AD from Caesarea Maritima (Palestine: I. Caesarea Maritima 229= CIIP no. 1477).

25The Syriac palaeography suggests an early dating, 4th century or the very beginning of the 5th cent. The Greek letter-forms (lunate sigma, the shape of the omega) are compatible with such a date.

Personal names

26The personal names (Table 1) are obviously meant to be close equivalents in the two languages and the vowels of the Greek help with the vocalization of the Aramaic (though the exact pronunciation of the Aramaic remains uncertain):

Table 1

Greek

Aramaic

Notes

Zoōra

Zeʿōrā

Apparently of Aramaic origin

Bia

Bīyā

Possibly of Greek origin

Boēthos

Boʾēthos

Common in Greek and here probably of Greek origin (note Greek case-ending in the Aramaic)

Inneos

Inī

Apparently of Aramaic origin (but provided with a Greek ending)

Zoōra is the best-attested of the four names in this inscription, paralleled in the broader area of Syria, including two instances from Kyrrhestike (see Yon, 2018, p. 160, and on parallels n. 107). It appears in the form Ζοώρας (commissioner of a funerary monument, Beroia/Aleppo (IGLSyr 1.194 of AD 294)); in the (genitive?) form Ζοώρα<ς> (in a funerary inscription from Zeugma (IGLSyr 1.98= SEG 26.1610 X.1; IGLSyr 2.661)); in the dative (in a funerary inscription from the region of Antioch (Kouaro) (IGLSyr 2.661 of AD 223)). It appears as a patronymic in IGLSyr 5.2509 of AD 167 from Tell Ameri in the region of Apamaea/Emesene (Χοσβα[νη](?) Ζοώρου); it appears also as Ζόρας in SEG 20.404 from Dura-Europos, 3rd century AD, and as Ζεώρου in SEG 40.1412, a funerary inscription from Baalbek (Heliopolis). In the text above we propose that the name Ζοώρα was treated as an indeclinable form. However, it may be that the letter-cutter simply omitted to add the sigma at the end of the word, in which case it should be read as Ζοώρα<ς>.
There are other interesting parallels to the name in Syriac literature and in Jewish Babylonian Aramaic (Yon, 2018, p. 160, n. 107). Abgar bar Zᵉʿōrā is named as one of the strategoi for the (Seleucid) year 618 in The Martyrdom of Šᵉmūnā and Gūryā (Burkitt, 1897, §1, 6; Yon, 2018, p. 186). Note also Mār Zᵉʿōrā after whom a church in Amīdā was named (Chronicle of Zuqnīn IV, 151-52 (Chabot, 1933, CSCO 104); Assemani & Assemani, 1926, III, 81). Yon classes the name in the category of those which have no religious connotation, i.e. could be borne by both Christians and pagans. We may also note Zaoras in Latin, frequently attested in the list of Palmyrene soldiers at Dura-Europos (P. Dura 98, 100 and 101. See Yon, 2018, pp. 110, 111, 113, 114).

27There is no doubt about the meaning of this Semitic name, “small”, as in Syriac (TS: 1144-5) and other branches of Aramaic (Elephantine, Nabataean, Jewish Palestinian, Christian Palestinian: Hoftijzer & Jongeling, 1995, pp. 337-338). A feminine version used for the name of a female appears in a Syriac inscription dated AD 243: zʿwrtʾ (Güler, 2014, p. 71).

28The personal name Bia is rare in Greek but is occasionally attested, e.g. on a funerary monument from Bithynia (in the genitive: Κάτωνος καὶ Βίας θυγάτηρ: IK Iznik 1335), but the name Βίας (nominative) is common in Greek inscriptions. See, e.g., its appearance in an inscription from Cilicia: Βίας Βαργαθέους, Zgusta, 1964, p. 120 §150-1.

29Boethos is widely attested as a Greek name. See Yon, 2018, p. 206-207 and n. 511, also 175: Vitus/Βῖτος/byṭws, bishop of Harran (4th cent.).

30With regard to the name Inneos, again the Greek gives us information on the pronunciation of the Aramaic and this extends also probably to the other Syriac occurrences of the name ʾny in the corpus of Edessan Aramaic inscriptions: Am3 and the recently discovered female statue from Harran (Önal & Desreumaux, 2019; and see Yon, 2019, p. 198). It might be linked to the Greek Ἔνεος, Ἐνναία, Ἐνναῖος (or Ἑνναῖος) and Ἔννης (Yon, 2018, pp. 166, 173, 175). Inneos itself is not otherwise attested in Greek but, if it is not a Semitic name provided with a Greek suffix, may derive through iotacism from the well-known Greek Αἰνέας, Αἰνείας or Αἴνειος.

31A remarkable detail is the fact that the name Zᵉʿōrā, not known elsewhere in early Syriac epigraphy, appears again in M 53 and also in one of the Kale Eteği tombs, though there the inscription is thought to be quite late in date (Desreumaux & Önal, 2017, p. 138, see below). This could be a coincidence, but one can at least speculate that this name indicates a family connection between the second Zᵉʿōrā (no patronym provided) and the first (son of Marbaytā [?] who himself clearly had sons). Tombs M 53, M 54 and M 55 appear to belong to a single enclosure forming a forecourt and separated by walls from adjacent tombs. The inscription in M 53 appears to be earlier than that of M 54 and it is possible that the one Zᵉʿōrā is a descendant of the other and that this is a family tomb complex.

Greek inscription from Kızılkoyun Tomb M 55 (and illegible Syriac inscriptions outside the tomb) (Figs. 3, 7, 8, 9)

Fig. 7: Syriac inscription on left-hand abacus of M 55

Fig. 7: Syriac inscription on left-hand abacus of M 55

B. Çetin and M. Demir

Fig. 8: Syriac inscription on right-hand abacus of M 55

Fig. 8: Syriac inscription on right-hand abacus of M 55

B. Çetin and M. Demir

Fig. 9: Greek inscription from the sarcophagus inside M 55

Fig. 9: Greek inscription from the sarcophagus inside M 55

B. Çetin and M. Demir

Syriac

32Tomb M 55 has an entrance which is designed to resemble a temple, on the templum in antis plan in the Corinthian order, with a decorated tympanum on a pediment supported by two columns. There are human heads (female?) on the capitals and barely legible Syriac inscriptions on the abacus of each: the writing on the left capital is marginally clearer and it may very tentatively be read as follows:

33dʿbd … [human head] lšʿd…..

34It will be obvious that there is little hope of light on this enigmatic inscription, though dʿbd appears to be the regular introductory formula meaning “which he made”, followed by a personal name. To the left of the human head we may have the name of the tomb owner. On the right-hand capital nothing at all can be read, though the presence of several Syriac letters is certain. (The inscribed areas are approximately 15 cm x 2 cm each side of each head.)

35To this kind of Syriac inscription there is, however, one good parallel in a Syriac inscription on the abacus of a capital from Harran published by Segal (As57; Segal, 1957; Laflı, 2016, p. 447, no. 2). In fact, the details of that column are very similar to those of M 55, to the extent of possibly having been made in the same workshop (see image in Segal, 1957, pl. II).

36Arcosolia are located on the south, north and east walls of the single burial chamber, with stone coffins below. In the middle of the chamber, there is an east-west oriented sarcophagus (length 225 cm, width 100 cm, height 100 cm) with a cover-stone which has undecorated corner acroteria. There is a panel forming the west side of this sarcophagus bearing an outward-facing Greek inscription.

Greek

37Dimensions: The flat panel on which the inscription is carved: height 40 cm (on right)-41 cm (on left), width 86 cm (top)-85 cm (bottom). Inscribed line-length (lines 1 and 2) 43.5 cm. Space between lines 1-1.5 cm.
Letter height: 4.5-6 cm.
Lettering: broken-bar alpha, lunate sigma and epsilon, curved omega.

38The inscribed lines are written on a raised panel.

1. ΟΑΝΑΓΕ vac. 17cm Α̣ΤΩΝ
2. ΕΝΘΑΚ vac. 17cm ΙΡΟΙΣ
vac. 67cm ΩΑΥΤΟΥ

Line 1
ΟΑΝΑΓΕ

39ὁ ἀναγε[ινώσκων] is a possible restoration, perhaps a form of ἀναγινώσκων, meaning “the one who perceives/reads”, which appears in various commemorative monuments from the Greek world. We might compare εὐτύχει ὁ ἀ[ναγεινώσκων] (“the reader is fortunate”: Dura-Europos VIII,2 96-7, n. 7, from Dura-Europos); cf. also μνησθῇ ὁ γράψας καὶ ὁ ἀναγεινώ[σ]κων ἐγὼ (“the writer and I the reader are reminded”: SEG 35.1536,A from Eleutheropolis, Palestine); μνη̣[σθῇ] ὁ γράψα̣[ς κα]ὶ ὁ ἀναγ̣ε̣[ινώ]σκων (“the writer and the reader is reminded”: YClS 14.1955.152,29 from Dura-Europos of the 3rd century AD); ὁ ἀναγιγνώσκων χαιρέτω (“let the reader salute”: KILyk I 11= MAMA 8.204 from Lykaonia). Other possibilities include ὁ ἀναγε[ννήσας], “he who has raised …” (see 1 Peter 1:3). However, the lack of epigraphical parallels and the absence of other signs of Christianity on this stone (there are no crosses) make this unlikely.

40If the first letter could be a theta with missing horizontal and the gamma an uneven tau, then we could have Θάνατε (“O, death!”), which is paralleled on grave epigrams from the Greek world. But this solution seems to be an unlikely one.

Α̣ΤΩΝ

41There are the traces of an alpha before the tau. Thus, we have a genitive plural ending or Α followed by a definite article. (On the implications of the traces of the alpha, see below.) There are endless possible restorations, including [τῶν λαμπροτ]άτων (cf. IG XIV 879 (Campania): Ἐλπίδιος ναύκληρος Συμμάχων τῶν λαμπροτάτων ἐνθάδε κῖται).

Line 2
ΕΝΘΑΚ

42ἔνθα κ[αὶ ἔνθα] is a possibility but more likely is ἔνθα κεῖται (“lies here…”), which is a funerary formula known, e.g., in SEG 8.270= CIIP no. 2484 from Gaza dated AD 449/50).

ΙΡΟΙΣ

43A dative plural seems to be the most likely explanation here (e.g. ἑταῖροις; καιροῖς; ὄνειροις), but another is the optative form [χα]ίροις (“you rejoice”: cf. KILyk I 15= MAMA 8.196 from Lykaonia).
Line 3

ΩΑΥΤΟΥ

44It appears that someone else is to be commemorated alongside the primary deceased person. Perhaps we should restore [σὺν τῇ συνβί]ῳ αὐτοῦ (“with his companion”: cf. ἐνθάδε κεῖται Ἀλέξανδρος Ἀνεμουριεὺς Ἰουδαῖος σὺν τῇ συνβίῳ αὐτοῦ, Heberdey and Wilhelm, Reisen in Kilikien 68,145 from Korykos in Cilicia) alternatively, [τῷ φίλ]ῳ αὐτοῦ is plausible.

  • 3 Indeed, Tracy’s assessment of the cutting process still allows for the possibility of preliminary s (...)

There is a seemingly uninscribed space between the Greek on the left and the right. It is possible that the letters in this blank space were added in with paint, but the traces of alpha in line 1 make it highly likely that the gap in all lines in the middle of the stone is the result of it being unfinished. There are certainly other examples in Greek epigraphy where what is extant on a stone seems to consist not of the final lettering but rather of a preliminary incision: see, for instance, SEG 31.1545 (OGIS 130; Egypt, ca. 143-142 BC) line 38, where a name seems to be incompletely incised. A better example is that of the catalogue of the thearodokoi of the Nemean Games and the Argive Heraia of 323/2 BC in which lines 25-32 of column B on the left-hand side of the stone appear to consist of preliminary incisions (Miller, 1988, pp. 148-49). Tracy, 1975, pp. 115-116 argues against the idea that preliminary sketches (e.g. in paint or graphite) were made of lettering of complete texts upon stone inscriptions from Athens. Tracy’s conclusions are, of course, based upon his study of lengthy Attic texts with small letters. However, in a text like the one we are publishing, with only a small number of words, in which line endings deliberately coincide with word-endings and the aesthetics of lettering (including visual articulation and symmetry) are important for the commemorative function of the inscription, it seems reasonable to suppose that the letter-cutter started at the end of each line, intending to fill in the rest after making preliminary incisions and before imposing a deeper cut to finalise each letter.3

45A very tentative reconstruction of lines 2-3:

ὁ ἀναγε[ινώσκων] …………… Α̣ΤΩΝ
ἔνθα κ[εῖται NAME καὶ τοῖς ἑτα]ίροις
………………..σὺν τῇ συνβί]ῳ αὐτοῦ.

“The reader …
Here lies NAME and his companions
……………………with his wife.”

The Greek and Syriac Inscriptions from Kale Eteği tombs

The Kale Eteği necropolis (Fig. 10)

Fig. 10: General view of Kale Eteği necropolis

Fig. 10: General view of Kale Eteği necropolis

J. Healey

46The Kale Eteği (= “foot of the Castle”) necropolis is located on the steep, rocky, limestone slope to the west of the citadel, which projects beyond the line of the western wall of the city for good strategic reasons (Sinclair, 1990, pp. 8-9). This necropolis has a surface area of 11,680 square meters and the slope is terraced, with rock-cut burial chambers set side by side. The first season of excavation took place in 2014 and the second season in August 2015. A total of 76 rock-cut burial chambers, built along the east and west of the Kale Eteği slope, were unearthed during the two seasons. Most were damaged due to their use as dwellings prior to clearance and expropriation.

47There are many dromos burial chambers, with a slope down to the door, though some lack this feature. The chambers follow the topography, so that there is no predominant direction of the burial chambers (unlike at Kızılkoyun) and it seems that no specific standards were applied to their plans and sizes. Square and rectangular plans were encountered and each chamber was of a different size, reflecting use as family graves and the number of individuals constituting the family. In some tombs the doorway that provided access to the tomb was closed with a rectangular monolithic block of stone and in others with a round stone (which could be rolled into place). In the tomb-chambers there are niches on the three sides of the tomb, with arched arcosolia, and some tombs have niches in the walls creating loculi for burials.

48These tombs appear to have been made in several different forms. The upper part of the dromos that provides access to the grave chambers was often covered with a vault, because the topography of the Kale Eteği meant that a vaulted entrance could prevent rainwater from entering the chambers. Some of the tombs contain mosaics with inscriptions set into them in Syriac. Others have frescoes (see below).

49As a result of the archaeological excavations and in the light of the architectural and archaeological findings, the burial chambers of the necropolis can be dated between the 2nd and 6th centuries AD.

List of the Inscriptions from Kale Eteği tombs

  • Tomb M 69 has a mosaic floor of typical early Edessan style, with four images of family members, with names beside each image (Uludağ & Önal, 2017, pp. 116-118; Desreumaux & Önal, 2017, pp. 132-134): dgdyʾ br brsmyʾ / ʾmʾ ʾntt gdyʾ sprʾ / ʾptwḥʾ br gdyʾ /mgdl ʾntt ʾptwḥʾ, “Belonging to Gadyā son of Barsamyā / Amā wife of Gadyā the scribe / Aptūḥā son of Gadyā / Magdal wife of Aptūḥā.” (Gadya mosaic)

  • Also from M 69 comes a broken tabula ansata bearing a Syriac inscription (Uludağ & Önal, 2017, pp. 116-118; Desreumaux & Önal, 2017, p. 135): it is certainly of much later date. ☩ ʾ[y]twhy qbrʾ hnʾ dsbʾ šmšʾ br ywḥnn wdʾlyʾ [….] ywsp […. ….]ʾ, “This tomb is that of Saba the deacon, son of John and of Elya […..] Joseph [….. …..]” (Inscription of Saba)

  • M 67 contains a mosaic inscription in Syriac published in Desreumaux & Önal, 2017, p. 137: dʿbd ʾksnts hnʾ byt ʿlmʾ lh wlbnwhy wlyrtwhy klhwn, “This is the house of eternity which Aksantos made for himself and his children and all his heirs”. The cross indicates that this is a Christian burial, though the traditional formula of the inscription is religiously neutral. (Xantos mosaic)

  • M 65: Also from Kale Eteği comes another Christian mosaic, that of the Abdu family (Ululdağ & Önal, 2017, p. 119; Desreumaux & Önal, 2017, 136): ☩ I […….. b]r ʿbdw br md/rywn ʿbdt ly byt ʿlmʾ hnʾ ly wlbny [w]lyrty ☩, “I … son of ʿAbdū son of Maryōn (?) made this house of eternity for myself, for me and for my children [and] for my heirs.” The name Maryōn is attested later as the name of a bishop of Šura in the time of Severus of Antioch (TS II, 2225). Again, the formula is not specifically Christian. (Abdu mosaic)

  • M 33 has two fresco-painted panels on arcosolia (see Uludağ & Önal, 2017, pp. 122-125 for photographs and minimal discussion). The text is studied in full below (1 and 2). (Akibsima and Patrōinos frescoes)

  • There is also an as yet unpublished Syriac inscription in a mosaic from M 5 (see Demir, 2019, pp. 259 and 270, plate 15 upper left).

Greek Inscriptions in Frescoes from Kale Eteği Tomb M 33

50The tomb contains two arcosolia. On the front of each arcosolium (left and right) there is a Greek text painted on the plaster which was used to render the stone structure.

Painted Greek Inscription on the Left arcosolium: Inscription of Akibsima (Fig. 11)

Fig. 11: Inscription on left arcosolium in tomb M 33

Fig. 11: Inscription on left arcosolium in tomb M 33

B. Çetin and M. Demir

51Painted on the left arcosolium of the tomb. Red painted letters, non-incised, on a white background within a painted representation of a squarish tabula ansata the dovetail handles of which are held by painted figures either side of it. Above the tabula is a Christian cross; below it, foliage and another Christian cross.

52Dimensions: Inscription area height 27 cm, width 40 cm; line length 39-40 cm, line height 3-3.5 cm; space between lines 1 cm. The border forming the tabula ansata is 3 cm thick.
Letter height: 3-3.5 cm
Lettering: The letters feature at times pronounced serifs; there are broken-bar alphas, lunate sigma, curved epsilon and omegas. The mus are inconsistent, sometimes featuring straight, at other times curved, verticals. There is some damage to the painting which means some letters are lost.
Previous Publication: Uludağ & Önal, 2017, pp. 122-125 no.1 (photographs.).
Date: 4th-6th century AD?

1. ΔΟΞΑΠΑΤΡΙΚΑΙΥΙΩΚΑΙΑΓΙ
2. ΩΠΝ̣ΕΥΜΑΤΙΕΛΠΙΣΗΜΩΝΑ
3. ΜΗΝΥ.ΕΡΜΝΗΜΗΣΑΚΙΒΣΙΜΑ
4. ΚΥΡΟΥ . . . ΜΑΡΟΥΘΑΠΕΝΘΕΡΑ
5. ΝΑΥΤΟΥ̣ΚΑΙΧΑΛΛΩΑΣΓΑΜΕ
6. ΤΗΝΑΥΤΟΥΚΑΙΚΥΡΟΥΚΑΙ,
7. ΣΑAKΟΥΥΙΩΝΑΥΤΟΥΚΑΙΚ
8. ΛΗΡΟΝΟΜΩΝΠΑΝΤΩΝ

δόξα Πατρὶ καὶ Υἱῷ καὶ Ἁγί-
ῳ Πν̣εύματι, ἐλπὶς ἡμῶν ἀ-
μήν. ὑ[π]ὲρ μνήμης Ἀκιβσιμα
Κύρου [καὶ] Μαρουθᾶ πενθερά-
ν αὐτοῦ̣ καὶ Χαλλωας γαμέ-
την αὐτοῦ καὶ Κυρου καὶ
Σαακοῦ ὑιῶν αὐτοῦ καὶ κ-
ληρονόμων παντῶν.

“Glory to the Father and the Son and the Holy
Spirit, our Hope, Amen.
In memory of Akibsima son of Kyros and
Maroutha, his mother-in-law,
and Challoa, his wife, and of Kyros and
Saakos, his sons, and all (his) heirs.”

  • 4 On the significance of the tabula ansata form, see above, A 2.

53The inscription is flanked by standing figures painted outside and supporting by hand the tabula ansata of the inscription (see Fig. 12).4

Fig. 12: Full view of left arcosolium in tomb M 33

Fig. 12: Full view of left arcosolium in tomb M 33

B. Çetin and M. Demir

54On the left is a figure of a woman, holding a Christian processional cross. On the right is another female figure, but the image is damaged. There are also crosses above and below the tabula. The imagery of the whole scene is unusual and demands the attention of and further research by specialists in Christian iconography (see description in Uludağ & Önal, 2017, pp. 122-123).

55The phrase δόξα Πατρὶ καὶ Υἱῷ καὶ Ἁγίῳ Πνεύματι is epigraphically well-attested in inscribed dedications and commemorative monuments in the area of Syria from the 4th to 6th centuries AD (IGLSyr 2.598, 4.1462, 4.1580, 4.1608, 4.1640, 4.1888; SEG 20.346). ἐλπὶς ἡμῶν is less well known but is paralleled in a plaque of grey marble from Attica of the 5th or 6th century AD (IG II13315).

56There is one odd linguistic feature: the apparent switch of case from genitive to accusative after ὑπέρ μνῆμης. This formula is usually followed by the genitive case: see, for instance, IGLSyr 5.2031 from Emesene of the 6th century AD (τοῦ ἁγίου Σεργίου ὑπὲρ μνήμης Βαραδάτου υἱοῦ Ἡλιοδόρου). The forms of the personal names in lines 3 to 5 (Ἀκιβσιμα, Μαρουθᾶ, Χαλλωας) which follow ὑπέρ μνῆμης may well be, as expected, in the genitive case; the same applies to Κύρου (line 6) and Σαακοῦ (line 7) and both ὑιῶν αὐτοῦ (line 7) and κληρονόμων πάντων (lines 7-8). However, the family terms which should be in the genitive in lines 4-5 and 5-6, i.e. πενθεράν (in apposition with Μαρουθᾶ) and γαμετήν (in apposition with Χαλλωας), appear in the accusative.

Personal names

57Akibsima appears to be an Aramaic male name, the first element being ʿQB, meaning “protect”. Yon describes this root as “très productive” in the naming tradition of Upper Mesopotamia with numerous variants in Greek and Latin and devotes a detailed study to the names which occur (Yon, 2018, pp. 84-88). The closest epigraphic parallel to the form occurring in our text is in a Greek inscription of 359/60 from Nisibis, where Ἀκεψύμα is named as the presbyter who built the baptistery (IK Estremo oriente 62; Jarry, 1972, pp. 242-243, no. 74; Keser-Kayaalp & Erdoğan, 2013, pp. 147-148). We may note also ʿAqebšimā (ʿqbšmʾ), the name of a martyr-bishop of Ḥānīṯā in Adiabene (Bedjan, 1891, pp. 351-396, known in Sozomen, Ecclesiastical History II, 13, as Ἀκεψιμᾶς). In Old Syriac we have less exact matches: ʿqybʾ (in a mosaic: Güler, 2014, p. 71) and the name of a citizen of Marcopolis, šmšʿqb, in P. Euphr. 18: v, 10, 16.

58Kyros is originally Persian, an Achaemenid royal name, in Syriac kūreš. It is well known in Greek epigraphical contexts, including Christian-era inscriptions from the area of Syria (IGLSyr 2.423 and 2.535). We note also Cyrus (qwrʾ), bishop of Edessa, at the end of the 4th cent. (Chronicon Edessenum xxxviii, xli) and Κῦρος /qwrʾ, bishop of Marcopolis, who was present at the Council of Ephesus in 431. And another Cyrus was the successor of Nonnus as bishop of Edessa in the late 5th (Chronicon Edessenum lxxi; Michael the Syrian, ed. Chabot 1905, III, 493). In Egypt, we have Cyrus the monk, a brother of Theodosius (Peeters, 1910, p. 56 § 238). See Yon, 2018, p. 207 n. 517.

59Maroutha is a well-known name in Syriac, though in our inscription it appears to be the name of a woman (“mother-in-law”), while in Syriac it is usually a male name, as in the cases of Marūṯā of Maypherqaṭ (4th/5th c. A.D.) and Marūṯā of Tikrit (d. 649). It is known in a Greek inscription from Caesarea Maritima (Palestine) of the 4th or 5th century AD (SEG 20.464= CIIP no. 1141).

Χαλλωας (not known in Greek inscriptions) is probably the genitive of Χαλλώα. A 5th-century bishop, Daniel, of Harran, had a mistress called Χαλλώα/klw: Yon 2018, p. 207, n. 514. The form may arise from spirantization: compare Χαλβας/βεις etc. from the Semitic root KLB: see Yon, 2018, p. 70, n. 438.
Saakos may be a Greek name, though it is not otherwise attested in Greek, and may derive from Isaakos, which is attested epigraphically (IK Estremo oriente 42). Simon Brelaud has drawn our attention to the Armenian Sahak or Šahak, also a deformation of the name Isaac. See Justi, 1895, pp. 272-273, 271 (Iranian Sag).

Painted Greek Inscription on the Right arcosolium: Inscription of Patroinos Zographos (Fig. 13)

Fig. 13: Inscription on right arcosolium in tomb M 33

Fig. 13: Inscription on right arcosolium in tomb M 33

B. Çetin and M. Demir

60Painted on the right arcosolium of the tomb. Red painted letters, non-incised, on a white background within a painted representation of a rectangular tabula ansata with an olive or palm branch to the left. The area to the right is damaged. Notably the general decorative features of the tabula are very different from those of the other arcosolium. More decoration has been included (though the human figures are omitted).

61Dimensions: Length of the four lines: 54 cm, 71 cm, 57 cm, 20 cm. The border forming the tabula ansata is 5 cm thick.
Letter-height: 4-5 cm.
Lettering: There are broken-bar alphas, lunate sigma, curved epsilon and omegas.
Previous Publication: Uludağ & Önal, 2017, pp. 122-5 no. 2; Güler, 2014, p. 36.

1. ΗΚΛΙΝΗΑΥΤΗΕΣΤΙΝ
2. ΠΑΤΡΩΙΝΟΥΖΩΓΡΑΦΟΥΚΑΡΡΗΝΟΥ
3. ΚΑΙΤΩΝΚΛΗΡΟΝΟΜΩΝ
4. ΑΥΤΟΥ

ἡ κλινὴ αὐτή ἐστίν
Πατρωΐνου Ζωγράφου Καρρήνου
καὶ τῶν κληρονόμων
………………αὐτοῦ.

“This is the burial-niche of Patrōinos (son of) Zōgraphos, a Harranian, and of his heirs.”

62The Υ at the end of line 2 is written above the ΝΟ, presumably owing to a lack of space. This is either a simple miscalculation by the painter or the result of the writing having been added some time after the tabula had been drawn.

63Here there are no linguistic difficulties. The purpose of the text was to identify the owner or occupier of the arcosolium where the text was written.

64The identification of the monument as a kline is unusual but epigraphically we have, for instance, the tomb of Thyoschontes Morphou of the late first century BC/first century AD (I.Napoli II 182= IG XIV 788). Other tombs with Syriac inscriptions in the same area have a similar reference to heirs (As7, As16, As59, Am2, Am3, Am7; Healey, 2006 and a mosaic dated AD 243: Güler, 2014, p. 71) and the practice of sharing a burial site with them is attested epigraphically elsewhere in the Greek world, including Mylasa in Caria (Mylasa 342= CIG 2709) and Telmessos in Lycia (TAM II 33).

Personal names

65The name Patrōinos is epigraphically attested elsewhere in the Greek world (Magnesia 162; IG XIV 2416,8; IGUR I 149 and IV 1638) and appears in a letter of Theodoret of Cyrrhus: Πατρώϊνος ὁ δρακωνάριος (PG 83, cols. 1231-2. Letters of Theodoret of Cyrrhus: Letter 59, To Claudianos).

66Zōgraphos too is attested epigraphically in the Greek world (e.g. in Phrygia: MAMA VI List 151,215; Diocaesarea: MAMA III 83).

67The ethnic Καρρηνός is not epigraphically attested to our knowledge. It was, however, known in antiquity: see Stephanos of Byzantium, Ethnika s.v. Κάρραι, offering also the alternative version Καρραῖος.

68Harran is nearby and it is noteworthy that Harran was a notorious centre of paganism during the period when Edessa began to boast of its Christianity (see Segal, 1963; Green, 1992). It appears that we here have a Harranian Christian, but the fact that his place of origin is indicated may suggest that he or his family was conscious of the fact that he was not originally from Edessa. He might have belonged to the Christian minority of Harranians who would have been more at home culturally in Edessa.

General

69Apart from the intrinsic interest of these well-provenanced inscriptions from Late Antiquity which help to fill out our picture of Edessa in this period, there are some other remarks which can be added on language-use in ancient Edessa.

70Their bilingual nature points to the bilingual culture of Edessa. Greek inscriptions and Syriac inscriptions are not uncommon, but here we have both languages in a bilingual text. Further, the two languages appear here to have equal status and this suggests that Edessan culture had moved on since the time of the Edessan Aramaic/Old Syriac inscriptions. Among the latter there is one bilingual (Mar Yakub: Drijvers & Healey, 1999, As62; note also Bm1), but otherwise Aramaic seems to predominate. It may be noted that our Greek texts contain some grammatical oddities (e.g. the apparent change of case from genitive to accusative after ὑπέρ μνῆμης in the Akibsima inscription (above, B 1) and the treatment of personal names as indeclinable in A 2b) which might reflect lack of full familiarity with standard Greek.

71After the annexation of Osrhoene to the Roman Empire, we know that Syriac continued to be used for legal texts (see P. Dura 28, P. Euphr. 18 and 19), in continuity with earlier Aramaic tradition and giving us evidence of the emergence of Syriac as a language of official documents. Greek, however, was probably becoming a prestige language (H language in sociolinguistic terms) to Syriac’s lower (at least less cosmopolitan) status (L). Semi-literates whose writing skill did not extend to Greek could add their signatures to Greek texts in Syriac script (see P. Euphr. 6-7, 9 and 10), since most legal documents were drawn up in Greek, and we can assume that Byzantine administration was conducted in Greek.

  • 5 See, for instance, the new Greek inscriptions published by Healey, Liddel, & Önal, 2020.

72Preference for Greek in certain contexts such as law and Byzantine administration, is unsurprising, while Syriac seems more prominent in religious and domestic spheres. The fact that the Bible was translated and promulgated in a Syriac version, as well as the emergence Syriac theological literature, suggest this preference in religion. By analogy with other areas within the Middle East (such as Palmyra), we might hypothesize that Aramaic would be on display in funerary contexts, though this stood alongside a tendency for some élite families to proclaim their Hellenized Christian culture by using Greek, even if this was not their normal vernacular. We can see this in the intrusion of Greek into the funerary sphere in the inscriptions published in this article. However, this use of Greek seems to have been limited to the 4th and 5th c., showing a tendency towards Hellenization which never became indispensable. Moreover, certain epigraphical details, such as the framing of inscriptions with the shape of the tabula ansata or the addition of crosses to the texts,5 are also reminiscent of the epigraphical habits of the Greek world in Late Antiquity.

73The other conclusion from this evidence is the important recognition that had not been suspected: the discovery of Christian inscriptions in the heart of the same Edessan necropoleis without any archaeologically-visible break. This expresses itself through the same language and script in which for at least a century and a half the culture of the élite had expressed itself, though the earlier, pagan tradition of depicting the dead with their families in floor-mosaics ceased. The funerary formulae remain identical, the script is regularized and stabilized, personal names persist while admitting new Greek names to the repertoire: members of the élite passed smoothly into Christianity without any disturbance, without pain, without rupture. The normalized script becomes the esṭrangelā script, that of the language of Scripture, which by this means conferred grammatical and literary status on the local dialect.

Haut de page

Bibliographie

Assemani, S. E., & Assemani, J. S. (1926). Bibliothecæ Apostolicæ Vaticanæ codicum manuscriptorum catalogus. Paris: Librairie Orientale et Américaine (reprint).

Bedjan, P. (ed.) (1891). Acta martyrum et sanctorum II. Paris/Leipzig: Harrassowitz.

Briquel Chatonnet, F. (2019). Writing Syriac: manuscripts and inscriptions. In D. King (Ed.), The Syriac World (The Routledge Worlds) (pp. 243-265). London/New York: Routledge.

Briquel Chatonnet, F., Desreumaux, A., & Moukarzel, J. (2008). Découverte d’une inscription syriaque mentionnant l’évêque Rabbula. In G. Kiraz (Ed.), Malphono w-Rabo d-Malphone: studies in honor of Sebastian P. Brock (Gorgias Eastern Christian Studies 3) (pp. 19-28). Piscataway, NJ: Gorgias Press.

Burkitt, F. C. (1897). Euphemia and the Goth: with the Acts of martyrdom of the confessors of Edessa (Text and Translation Society, London, 2). London: Williams and Norgate (reprint Amsterdam: Philo, 1981).

Çetin, B (2017). Şanlıurfa Kızılkoyun ve Kale Eteği Nekropolü Kutarma Kazı ve Temizlik Çalışmaları. In A. Özme (Ed.), 26. Müze Kurtama Kazıları Sempozyumu. 19-20 Nisan, 2017 Hatay (pp. 29-44). Ankara: T. C. Kültür ve Turizm Bakanlığı.

Chabot, J.-B. (1905). Chronique de Michel le Syrien patriarche jacobite d’Antioche (1166-1199) Éditée pour la première fois et traduite en français III. Paris: Ernest Leroux.

Chabot, J.-B. (1920). Chronicon ad annum Christi 1234 pertinens I (CSCO 81, Script. Syri. 36). Louvain: Secrétariat du CorpusSCO.

Chabot, J.-B. (1927). Incerti auctoris Chronicon Pseudo-Dionysianum vulgo dictum I (CSCO 91, Script. Syri. 43). Paris/Louvain: Secrétariat du CorpusSCO.

Chabot, J.-B. (1933). Incerti auctoris Chronicon Pseudo-Dionysianum vulgo dictum II (CSCO 104, Script. Syri. 53). Paris/Louvain: Secrétariat du CorpusSCO.

Cureton, W. (1864). Ancient Syriac Documents: relative to the earliest establishment of Christianity in Edessa and the neighbouring countries. London/Edinburgh: Williams and Norgate.

Demir, M. (2019). Şanlıurfa Kızılkoyun ve Kale Eteği Nekropolü Kurtama Kazı ve Temizlik Çalışmarı. In M. Önal, S. İ. Mutlu, & S. Mutlu (Eds.), Harran ve Çevresi Arkeoloji (pp. 251-270). Şanlıurfa: Şurkav Yayınları.

Desreumaux, A. (2009). Comment peut-on écrire en syriaque ou Des problèmes du scribe devant sa page blanche. In Ch. Batsch & M. Vârtejanu-Joubert (Eds.), Manières de penser dans l’Antiquité méditerranéenne et orientale. Mélanges offerts à Francis Schmidt par ses élèves, ses collègues et ses amis (pp. 105-126). Leiden, Brill.

Desreumaux,, A. & Önal, M. (2017). The Translation of Syriac Inscription(s) of New Mosaics found in Şanlıurfa. In M. Önal (Ed.), Urfa-Edessa Mozaikleri (pp. 132-41). Şanlıurfa: Şanlıurfa Büyükşehir Belediyesi, Kültür ve sosyal işler daire başkanlığı.

Drijvers, H. J. W., & Healey, J. F. (1999). The Old Syriac Inscriptions of Edessa and Osrhoene: texts, translations and commentary (Handbuch der Orientalistik I/42), Leiden, Brill.

Duval, R. (1892). Histoire politique, religieuse et littéraire d’Édesse jusqu’à la première croisade. Paris: Extrait du Journal Asiatique 18-19 (1891-92). (reprint Amsterdam: Philo).

Green, T. M. (1992). The City of the Moon God: Religious Traditions of Harran (Religions of the Graeco-Roman World 114). Leiden/New York/Cologne: E. J. Brill.

Güler, S. E. (2014). Şanlıurfa Yazıtları (Grekçe, Ermenice ve Süryanice). Istanbul: Arkeoloji ve Sanat Yayınları.

Healey, J. F. (forthcoming [2021]): Minimal Literacy: some evidence from the earliest Syriac documents. [to appear in a Festschrift]

Healey, J. F., Liddel, P. P., & Önal, M. (2020). New Greek Inscriptions from Harran Castle. Zeitschrift für Papyrologie und Epigrafik, 216, 133-46.

Hillers, D. R., & Cussini, E. (1996). Palmyreme Aramaic Texts. Baltimore, MD/London: Johns Hopkins University Press.

Hoftijzer, J., & Jongeling, K. (1995). Dictionary of the North-West Semitic Inscriptions (Handbuch der Orientalistik I/XXI.1-2). Leiden/New York/Cologne: E. J. Brill.

Jarry, J. (1972). Inscriptions syriaques et arabes inédites du Ṭūr ʿAbdīn. Annales Islamologiques, 10, 207-50.

Justi, F. (1895). Iranisches Namenbuch. Marburg: N. G. Elwert.

Keser-Kayaalp, E., & Erdoğan, N. (2013). The Cathedral Complex at Nisibis. Anatolian Studies, 63, 137-154.

Laflı, E. (2016). Five Old Syriac Inscriptions from the Museum of Şanlıurfa (Ancient Edessa). Aram, 28, 443-451.

Leatherbury, S. V. (2018). Framing Late Antique Texts as Monuments: The Tabula Ansata between Sculpture and Mosaic. In A. Petrovic, I. Petrovic, & E. Thomas (Eds.), The Materiality of Text: Placement, Perception, and Presence of Inscribed Texts in Classical Antiquity (Brill Studies in Greek and Roman Epigraphy 11) (pp. 380-404). Leiden: E. J. Brill.

Leatherbury, S. V. (2019). Inscribing Faith in Late Antiquity: between Reading and Seeing (Image, Text, and Culture in Classical Antiquity). London: Routledge.

Luther, A. (1997). Die syrische Chronik des Josua Stylites (Untersuchungen zur antiken Literatur und Geschichte 49). Berlin/New York: de Gruyter.

Miller, S. G. (1988). The Theorodokoi of the Nemean Games. Hesperia, 57, 147-63.

Negev, A. (1971). A Nabatean Epitaph from Trans-Jordan. Israel Exploration Journal, 21, 50-52.

Önal, M. (2017). Urfa-Edessa Mozaikleri. Şanlıurfa Büyükşehir Belediyesi, Kültür ve sosyal işler daire başkanlığı.

Önal, M., & Desreumaux, A. (2019). The Fragment of a Woman[’s] Statue with Syriac Inscription [was] Found in Harran Excavation. In M. Önal, S. İ. Mutlu, & S. Mutlu (Eds.), Harran ve Çevresi: Arkeoloji (pp. 241-49). Şanlıurfa: Şurkav Yayınları.

Önal, M., Ercan, M., Desreumaux, A., & Dervişoğlu, N. (2013). Restore Edilen İki Adet Edessa Mosaiği 2011 / Two Edessa Mosaics which were restored in 2011. Journal of Mosaic Research, 6, 9-21.

Payne Smith, R. (1879-1901). Thesaurus Syriacus. Oxford: Clarendon Press.

Peeters, P. (1910). Bibliotheca Hagiographica Orientalis (Subsidia Hagiographica 10), Brussels: Bollandists.

Sabbagh, R., Ayash, F., Balty, J., Briquel Chatonnet, F., & Desreumaux, A. (2008). Le martyrion Saint-Jean dans la moyenne vallée de l’Euphrate. Fouilles de la Direction Générale des Antiquités à Nabgha au nord-est de Jarablus (Documents d’archéologie syrienne XIII). Damascus: Ministère de la Culture, Direction Générale des Antiquités et des Musées.

Sachau, E. (1882). Edessenische Inschriften. ZDMG, 36, 142-67.

Segal, J. B. (1957). Two Syriac Inscriptions from Harran (Appendix by Donald Strong). Bulletin of the School of Oriental and African Studies, University of London, 20 (Studies in Honour of Sir Ralph Turner), 513-522.

Segal, J. B. (1963). Edessa and Harran: an inaugural lecture delivered on 9 May 1962. London: School of Oriental and African Studies, University of London.

Segal, J. B. (1970). Edessa: “The Blessed City”. Oxford: Clarendon Press.

Sinclair, T. A. (1990). Eastern Turkey: an Architectural and Archaeological Survey IV. London: Pindar Press.

Sokoloff, M. (1990). A Dictionary of Jewish Palestinian Aramaic. Ramat-Gan: Bar Ilan University Press.

Sokoloff, M. (2002). A Dictionary of Jewish Babylonian Aramaic. Ramat-Gan: Bar Ilan University Press.

Tracy, S. V. (1975). The Lettering of an Athenian Mason (Hesperia Supplement XV). Princeton, NJ: American School of Classical Studies at Athens.

Trombley, F. R., & Watt, J. W. (2000). The Chronicle of Pseudo-Dionysius the Stylite (Translated Texts for Historians 32). Liverpool: Liverpool University Press.

Uludağ, C., & Önal, M. (2017). Edessa, Kale Eteği Mozaikler. In Önal, M. (Ed.), Urfa-Edessa Mozaikleri (pp. 116-125). Şanlıurfa Büyükşehir Belediyesi, Kültür ve sosyal işler daire başkanlığı.

Üney, S. (2015). Haleplibahçe Mosaikleri (Doctoral Thesis Ege University, İzmir). İzmir: T.C. Ege Üniversitesi, Sosyal Bilimler Enstitüsü, Klasik Arkeoloji Anabilim Dalı.

Yadin, Y., Greenfield, J. C., Yardeni, A., & Levine, B. A. (2002). The Documents from the Bar Kokhba Period in the Cave of Letters. Hebrew, Aramaic and Nabatean-Aramaic Papyri (Judean Desert Studies) (2 vols.). Jerusalem: Israel Exploration Society/Hebrew University/Shrine of the Book.

Yon, J.-B. (2018). L’Histoire par les noms. Histoire et onomastique, de la Palmyrène à la Haute Mésopotamie romaines. Beirut: Presses de l’IFPO.

Zgusta, L. (1964). Kleinasiatische Personennamen (Československá Akademie Věd, Monografie Orientálního ústavu ČSAV 19). Prague: Verlag der Tschechoslowakischen Akademie der Wissenschaft.

Haut de page

Annexe

Abbreviations

Abbreviations referring to Greek inscriptions are those used on the Packard Humanities Institute website: https://epigraphy.packhum.org

Further abbreviations:

CIIP= Corpus Inscriptionum Iudaeae/Palaestinae

CIS= Corpus Inscriptionum Semiticarum

CSCO= Corpus Scriptorum Christianorum Orientalium

IGLSyr= Inscriptions grecques et latines de la Syrie (IFPO), Beyrouth

PAT= Hillers & Cussini 1996

PG= Patrologia Graeca

TS= Payne Smith 1879-1901

Haut de page

Notes

1 See also discussion in Luther, 1997, pp. 154-55 and n. 211; Trombley & Watt, 2000, p. 25 n. 122.

2 For the depiction of tabulae ansatae as frames in order “to challenge the two-dimensionality” of the epigraphical medium, see Leatherbury, 2018, p. 399; cf. Leatherbury, 2019, pp. 93-97. See also below, B 1 and 2 from M 33.

3 Indeed, Tracy’s assessment of the cutting process still allows for the possibility of preliminary sketching of individual letters (cf. Tracy, 1975, p. 125, describing an experiment by a modern mason: “No attempt was made to complete a furrow in one cutting, but instead the chisel was driven over the surface three or four times....”).

4 On the significance of the tabula ansata form, see above, A 2.

5 See, for instance, the new Greek inscriptions published by Healey, Liddel, & Önal, 2020.

Haut de page

Table des illustrations

Titre Fig. 1: Map of Edessa
Crédits J. Healey
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-1.png
Fichier image/png, 82k
Titre Fig. 2: General view of the Kızılkoyun necropolis
Crédits J. Healey
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-2.tif
Fichier image/tiff, 704k
Titre Fig. 3: General view of tombs M 53, M 54 and M 55
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-3.jpg
Fichier image/jpeg, 73k
Titre Fig. 4: General view of the arcosolium with inscription in M 53
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-4.jpg
Fichier image/jpeg, 124k
Titre Fig. 5: The Syriac/Edessan Aramaic inscription in M 53
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-5.jpg
Fichier image/jpeg, 268k
Titre Fig. 6: Bilingual inscription from outside tomb M 54
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-6.jpg
Fichier image/jpeg, 84k
Titre Fig. 7: Syriac inscription on left-hand abacus of M 55
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-7.jpg
Fichier image/jpeg, 68k
Titre Fig. 8: Syriac inscription on right-hand abacus of M 55
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-8.jpg
Fichier image/jpeg, 102k
Titre Fig. 9: Greek inscription from the sarcophagus inside M 55
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-9.jpg
Fichier image/jpeg, 81k
Titre Fig. 10: General view of Kale Eteği necropolis
Crédits J. Healey
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-10.jpg
Fichier image/jpeg, 59k
Titre Fig. 11: Inscription on left arcosolium in tomb M 33
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-11.jpg
Fichier image/jpeg, 131k
Titre Fig. 12: Full view of left arcosolium in tomb M 33
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-12.jpg
Fichier image/jpeg, 86k
Titre Fig. 13: Inscription on right arcosolium in tomb M 33
Crédits B. Çetin and M. Demir
URL http://journals.openedition.org/anatoliaantiqua/docannexe/image/1384/img-13.jpg
Fichier image/jpeg, 77k
Haut de page

Pour citer cet article

Référence papier

Bekir Çetin, Müslüm Demir, Alain Desreumaux, John Healey et Peter Liddel, « New Inscriptions in Aramaic/Early Syriac and Greek from the Cemeteries of Edessa »Anatolia Antiqua, XXVIII | 2020, 119-141.

Référence électronique

Bekir Çetin, Müslüm Demir, Alain Desreumaux, John Healey et Peter Liddel, « New Inscriptions in Aramaic/Early Syriac and Greek from the Cemeteries of Edessa »Anatolia Antiqua [En ligne], XXVIII | 2020, mis en ligne le 01 décembre 2022, consulté le 12 mars 2026. URL : http://journals.openedition.org/anatoliaantiqua/1384 ; DOI : https://doi.org/10.4000/anatoliaantiqua.1384

Haut de page

Droits d’auteur

CC-BY-SA-4.0

Le texte seul est utilisable sous licence CC BY-SA 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont susceptibles d’être soumis à des autorisations d’usage spécifiques.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search