Lemaire A. and Sass B. The Mortuar PDF

The Mortuary Stele with Sam’alian Inscription
from Ördekburnu near Zincirli
André Lemaire and Benjamin Sass
For more than a century, the alphabetic inscription on the early-first-millennium Ördekburnu
stele has defied decipherment. In 1915, Lidzbarski labeled the language of the hardly legible text
non-Semitic, and most specialists have followed suit. We present a new collation of the stele, with
a drawing, and a partial transliteration and translation of the Sam’alian text.
Introduction The stele was pointed out to von Luschan at Ördek-
F
burnu in 18882 and was subsequently removed to the
or more than a century, the alphabetic inscription Archaeological Museum in Istanbul, where it has been
on the early-first-millennium Ördekburnu stele, on display since (Inv. 7696). For quite some time before
defying decipherment, has been considered a lost 1888, the local people had been using the stele as a sur-
cause for West Semitic studies. In 1915, the language of face for felt-making, damaging the relief and the inscrip-
the hardly legible text was labeled non-Semitic by the great tion. This led to uncertainty about the reading and even
Semitist Mark Lidzbarski, and most specialists have fol- the language when the stele was studied. The original site
lowed suit. Even the few who have judged the language to of the stele (a tomb or mortuary chapel?) is unknown—
be Semitic, probably Aramaic, have not advanced much Ördekburnu itself or another place, in the latter case
beyond pointing out a handful of words, divine names in presumably not far away (see “Conclusions: The Stele in
particular. In 1932, Helmut Bossert put his finger on the Sam’alian Geography and History”).
solution, but he let it slip, and his findings have practically Ördekburnu (“Duck’s Bill”) is a promontory, crossed
been forgotten. In the following, we present a new collation on the east by a tunnel of the Aleppo railway,3 18 km
of the stele, with a drawing, and a partial transliteration south of Zincirli, 10 km south of İslahiye and 5 km east
and translation of the Sam’alian text. of Altınüzüm (Figs. 3–7).4 It forms part of Katranlıdağ
(“Tarry Mountain”),5 the ridge separating here the
The Stele and Its Find-Spot western section of the Rift Valley from the eastern

(Karasu) section (Fig. 5). In Koldewey’s map of 1890–
This trapezoidal basalt stele (Figs. 1, 2), its top broken 1891 (Fig. 6), Ördekburnu and Katranlıdağ are marked
away, now measuring 120 × 54 × 21 cm, bears on its front as two parts of the ridge. Amir Fink, during a visit in
a table scene above a nine-line West Semitic alphabetic July 2008, verified that Ördekburnu is the name by which
inscription, both in relief. A tenth line, 80 cm long, con-
stituting the end of the text, is carved in relief from bot-
tom to top on the left side of the stele.1 On the mortuary
Sendschirli 4: pl. 55 bottom; Orthmann 1971: Zincirli G/1; Gilibert
role of the monument, see “The Table Scene . . .” below. 2011: 91, 219).
2 During the first excavation season at Zincirli; cf. Sendschirli 4:
1 In our side photograph of the stele, Fig. 1:b2, the view is partly 330, and caption to fig. 239 on p. 329.
obstructed by a sphinx orthostat from Zincirli exhibited nearby (cf. 3 In 1888, when von Luschan learned about the stele, the railway
line had not yet been constructed.

4 In Fig. 3, the Ördekburnu promontory is located left of the “Bilen-
André Lemaire: 21 bis Ave de Stalingrad, 91120 Palaiseau, ler” caption, itself wrongly placed: Bilenler village is situated just south
France, [email protected] of Ördekburnu, visible in Figs. 3 and 4. Altınüzüm (former name
Aydınoğlu) is a town on the İslahiye–Antakya road.
Benjamin Sass: Institute of Archaeology, Tel Aviv 5 In Google Earth, the label “Katranlıdağ” figured on 25 April 2010
University, P.O.B. 39040, Ramat Aviv, Tel Aviv 69978, Israel, about a kilometer southwest of the Ördekburnu promontory, in the
[email protected] middle of the plain—evidently not its right place.
© 2013 American Schools for Oriental Research. BASOR 369 (2013): 57–136.
This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

All use subject to JSTOR Terms and Conditions
58 LEMAIRE AND SASS BASOR 369
Fig. 1. Photographs of the stele. (a) front: 2920; (b1) side: unnumbered Museum photo; (b2), 2912; (c) three-quarter view: 605 (Istanbul Archae-
ological Museums and authors).
the local people now call the promontory.6 All this is not tile little plain, bordered by the Karasu on the east . . .”7
easily reconciled with von Luschan’s somewhat inexact (Sendschirli 4: 329).
localization: “Twelve km south of Zincirli . . . Between Concerning the Karasu, its main channel flows
two rocky elevations, the Katranlıdağ (Katranlidagh) several kilometers to the east, on the other side of the
and Ördekburnu (Oerdek-Burunu), lies a flat, very fer- Katranlıdağ ridge, but a branch, now transformed into
a canal, runs at the northwest foot of the Ördekburnu
promontory; one may hope that in 1888 this branch was
6 We are grateful to Amir Fink for this and several other valu-
able tips. Same in Google Earth, 25 April 2010—“Ördekburnu Tepesi, 7 The sentence ends: “. . . where some ten or twelve Kurdish families
İslahiye, Turkey.” pitch their tents every winter, purportedly for centuries.”

2013 THE MORTUARY STELE FROM ÖRDEKBURNU 59
While the course of Sam’al’s southern border is today

a matter of conjecture, there is little doubt that Ördek-
burnu lay within; not only do the language and content
of the text testify to this, but also the geographical logic:
as the sculpture quarry of Yesemek—some 25 km south
of Zincirli as the crow flies (Duru 2004)—belonged to
Sam’al, Ördekburnu probably did also.
Earlier Studies
Attempts at a decipherment by Peiser (1898) and von

Luschan9 were followed by Mark Lidzbarski, who col-
lated the inscription in Istanbul in September 1909 and
May 1910 (ESE 3: 192, 193), partly by candlelight in a
darkened room (ESE 3: 193), and presented as best as he
could a transliteration into square Hebrew letters (ESE 3:
197). He had three photographs taken of the front (ESE 3:
pls. 13–15), yet the bad state of preservation deterred him
from preparing a facsimile drawing.10 On the other hand,
Lidzbarski was the first to notice the 10th line, of which
he published a transliteration and a drawing (the latter,
our Fig. 8), but no photograph. As he could not decipher
the 10 lines, Lidzbarski proposed that their language was
Anatolian, with only a sprinkle of Semitic words,11 and
others followed him (RES 3: 1786 [pp. 292–95]; Sayce
1927: 713–14; Przeworski 1928: 233; Dhorme 1930: 28,
n. 1;12 Friedrich 1932: 38).13 Returning to Lidzbarski, it
is evident that he reached the decision about the Ana-
tolian language only reluctantly. This may explain why
he sought ʾnk at the beginning of line 1 (ESE 3: 197–98),
read nbkyn in the end of line 4 and beginning of line 5
(p. 200), and understood much of line 10 as Semitic af-
ter all (ESE 3: 201) and, what is more, recognized the
9 Von Luschan reported that “Unabhängig von Peiser habe ich
schon an Ort und Stelle Barrekub und melek ya/di lesen zu können
geglaubt, aber ich fürchte, daß das bloße Suggestion war . . .” (Send-
schirli 4: 330). “Barrekub” could have been rkbʾl in line 5 or 7, or both;
“melek ya/di”—is probably mlky in line 9.
10 “Die Überreste der Zeichen vermischen sich hier mit den Uneben-
located in accordance with von Luschan’s description. heiten des Grundes. . . . Eine Zeichnung herzustellen schien mir nicht
Finally, Katranlıdağ in Luschan’s report clearly pertains ratsam. Diese wäre für den größten Teil der Inschrift ganz subjektiv und
schlösse eine selbstständige Mitarbeit anderer aus” (ESE 3: 192, 193).
to a localized topographical feature—a “rocky elevation”
Even while a measure of subjectivity is undeniably there, Lidzbarski’s
on a par with Ördekburnu. Had he an adjacent promon- appraisal may have been unduly pessimistic; after all, exactly the same
tory in mind, referring to it by the name of the entire criticism could be voiced against his published transliteration.
ridge? The little plain between Ördekburnu and the next 11 “[E]in Denkmal . . . das alphabetisch geschrieben, aber in einer
promontory to the north may fit the above role:8 Criss- nichtsemitischen Sprache abgefaßt ist” (ESE 3: 192). “Es fragt sich nun
welche Sprache hier vorliegt, in die die semitischen Brocken eingebettet
crossed by canals today (replacing the erstwhile branches
sind” (p. 202). On pp. 205–6, Lidzbarski tried to find Lycian compar-
of the Karasu), this plain probably matches Luschan’s isons. And he summarized: “Die Sprache der Inschrift ist im Wesent
designation as “very fertile” better than its counterpart lichen nichtsemitisch, sie gehört wahrscheinlich zur kleinasiatischen
to the south. Gruppe” (p. 206).
12 “[L]e texte en reste très obscur et l’on hésite même sur la nature
du dialecte représenté. . . .”
8 We have not learned the specific name of the latter promontory. 13 Friedrich merely repeated Lidzbarski.

Fig. 2. Drawing of the stele (by Rodica Penchas and Yulia Gottlieb guided by the authors).

Fig. 3. View of Ördekburnu (Google Earth, 22 April 2011, © Google).

Fig. 4. View from Ördekburnu to the south, with triangulation point (photo by Amir Fink, 2008).
benedictory (or maledictory) character of line 10, with and Lidzbarski’s view about the non-Semitic language—
which we agree.14 at any rate, the ambiguity concerning the language—was
Helmut Bossert studied the stele ca. 1930 (cf. Bossert maintained by most specialists, even Bossert himself,
1932: 35–36, 87) and sketched the correct letter sequence as it turns out.18 Among the others who followed suit
of the best-preserved part of the text—line 8, with parts we may mention Cross and Freedman (1952: 21, n. 5);19
of lines 5–7 (Fig. 9), accompanying his drawings by a Koopmans (1962: Vol. 1: 22);20 KAI;21 Naveh (1970:
transliteration, though not by a translation:15 7–8);22 Meriggi (1975: 219–20);23 Gibson (TSSI 2: 159);24
Rosenkranz (1978: 9, “kaum semitisch”); Fitzmyer and
5•lrkbʾl• . . . 6•lkbb• (p. 35, n. 2), 7•brkbʾl•šʾ Kaufman (1992: 333, Appendix A: Aramaica?); and
8yn•lym•bkbb•šʾyn•lym• (p. 87).16 Dion (1997: 102).25 But, with the notable exception of
Bossert, most of these authors hardly studied the inscrip-
Bossert’s drawing and transliteration alone should tion themselves; predominantly they were just echoing
have made it clear to everyone that the language of the Lidzbarski. Thus, also Tropper (1993: 6), in his thorough
Ördekburnu inscription is West Semitic, but somehow
they did not. Bossert’s article was practically forgotten,17 18 “Vgl. Lidzbarski . . . 1915. . . . Über dem Relief Götterideo-
gramme in hethitischer Hieroglyphenschrift. Unter dem Relief eine

14 “[D]er Sinn der Zeile wird wohl sein: wer auslöschen sollte den Inschrift in aramäischer Schrift, aber höchstwahrscheinlich in hiero-
Namen des Ḫaian, sei verwünscht beim Gotte Šemeš. Weniger wahr glyphenhethitischer Sprache” (Bossert 1942: 76, no. 957).
scheinlich steht eine Benediktion da: wer gedenken sollte den Namen 19 “The text is very obscure. . . .”
des Ḫaian (vgl. ‫ יזכר אשם פנמו‬Had. 21), sei gesegnet (oder ähnlich) 20 “. . . vielleicht semitisch, aber mit fremden (hethitischen?)
beim Gotte Šemeš.” We do not share the readings Ḫaian (kyn) and Einflüssen.”
Šemeš (šmš), however. 21 No KAI entry, possibly no mention.
15 Bossert was principally interested in the goddess Kubaba, whose 22 “[I]t is difficult even to ascertain the language in which it is
name, written kbb, he read correctly in lines 6 and 8. Lidzbarski (ESE written.”
3: 199) recognized one of the two kap-bet-bet sequences, but not its 23 “. . . un’iscrizione in caratteri semitici antichi del Xo o IXo sec.
meaning (see n. 97). a. C., ma in una lingua, in cui i semitisti riconoscono solo pochi el-
16 Bossert’s only slip was his overlooking the waw preceding bkbb. ementi semitici e pensano quindi a una lingua non semitica, . . .”
17 Among the proponents of Aramaic cited below, Landsberger 24 The Ördekburnu inscription (“Cilicia; 9 cent.”) is first in the list
(1948: 49) mentioned Bossert 1932 in passing in connection with his of “Texts not included in the present volume.”
discovery of Kubaba’s name in the inscription, but not with regard to 25 “On n’a pu jusqu’à présent ni traduire l’inscription . . . , ni en
the language of the text. identifier la langue.”

2013

THE MORTUARY STELE FROM ÖRDEKBURNU

Fig. 5. F. Winter and F. von Luschan’s 1888 map, with permission, Vorderasiatisches Museum, Staatliche Museen zu Berlin, reproduction Olaf M. Teßmer (first published in Wartke 2005: 13).
63
Fig. 6. Koldewey’s map (Sendschirli 1, at end).

Fig. 7. Map by Steven D. Batiuk, 2012.

Fig. 8. Lidzbarski’s drawing of line 10 (ESE 3: 194).
Fig. 9. Bossert’s partial drawing of lines 5–8 (1932: 36, 87).
treatment of the other Zincirli inscriptions: although he (1987: 171–72), and Lipiński (2000: 233 and n. 6) fol-
did cite Bossert’s paper, he labeled the language of the lowed Landsberger.29
Ördekburnu text “possibly Luwian,” excluding it from
his discussion. Others—e.g., Bowman (1948: 71, n. 27)26 Our Study
and, in the same year, Landsberger (1948: 49 with n. 124,
64 with n. 166)—rightly regarded the language as West In the autumn of 1980, one of us (Lemaire) studied
Semitic, probably Aramaic, on account of the words the stele in the Istanbul Museum for the first time and
and divine names that Lidzbarski was able to identify.27 took a paper squeeze (Fig. 10). He was subsequently sent
Similarly Yadin (1970: 210).28 Voos (1986: n. 381), Sader photographs by the museum, four of the front (two of
them in Fig. 11) and one of the vertical line on the side
(Fig. 1b1). The Semitic, more specifically Sam’alian, lan-
26 “The much-debated Oerdek-burnu stela, so damaged as to be
guage of the inscription became immediately apparent
almost illegible . . . has been called non-Semitic due to difficulty in ob-
taining a clear translation. Where it is legible, it compares favorably (see Lemaire 2001: 187).30 Our joint expedition material-
with the other ‘Old Aramaic’ material, and it is certainly written in the ized after Sass, too, spent some time in front of the stele,
same type of old Semitic alphabet.” in September 2007.
27 “Die auf die Autorität von Lidzbarski . . . zurückgehende, un-
We studied the inscription at the Istanbul Archaeo-
widersprochen gebliebene Ansicht, dass diese Inschrift in einer nicht
logical Museum on 21–22 April 2008, accompanied by
semitischen Sprache (wofür nur Luisch in Frage käme) abgefasst sei,
ist aufzugeben . . . : soweit deutlich lesbare Wörter vorliegen, sind sie Ms. Evren Kıvançer of Istanbul, whose involvement con-
aramäisch, so alle sechs zweifelsfrei lesbaren Wörter der Randzeile tributed a great deal to the success of the expedition. In
[these include šmš ʾlh, which we contest] (ʾşm „Name,” hʾ „jener“ . . . the absence of Dr İsmail Karamut, the Museum’s director,
sind typisch für Had. und Pan.) . . . Die Inschrift hat als Grabinschrift who on 15 January 2008 had granted us permission to
zu gelten . . . Die Erwähnung von drei, bzw. vier Göttern (Rakkab-El,
study and photograph the stele for publication, we were
Kubaba, Şamaş und vielleicht Şarma in Z. 3) . . . spricht gegen die Auf-
fassung als Weihinschrift” (Landsberger 1948: 64 with n. 166). Let us
add in passing that the 10 lines of the inscription constitute the entire 29 Lipiński (2000: 233 and n. 6) was moreover the only one who
West Semitic text; with Landsberger (1948: n. 166), we wonder what tried to elaborate on Lidzbarski’s limited decipherment as Semitic, if
led Przeworski (1928: 233–34) to believe in the existence of more writ- with mixed results.
ing all around. 30 “En fait, même si son état ne permet pas d’en proposer une tra-
28 “Even though the reading of the inscription is most difficult, duction suivie, ce qui a déjà été lu par Lidzbarski et ce qu’on peut en lire
there is no doubt that in line 5 there is written ‫לרכבאל אלה‬, and in line à un examen attentif de l’original et des photographies suggèrent plutôt
7 there is possibly ‫)!( ברכבאל‬.” d’y voir une inscription sam’alienne.”

Fig. 10. Lemaire’s squeezes, 1980, images reversed: (a) front inscription; (b) side inscription, right; (c) side inscription, left (photos by the authors).
greatly helped by Mrs. Zeynep Kızıltan, Deputy Director. quired. That is, the nearly pristine Katumuwa inscription
Our warm thanks to them. could be transliterated right away, whereas our greatest
In Istanbul, in front of the original, we did a first investment of time and energy has gone into ascertain-
drawing of the inscription, subsequently improving and ing, and then documenting, the material aspect of the
completing it in Tel Aviv (see “Identification . . .” below). letters of the much-worn Ördekburnu inscription, con-
Almost at the same time that we were working on sidered undecipherable for so long. Once this task was
this inscription, the stele of Katumuwa was discovered at accomplished (in fact even before), we realized how
Zincirli and was promptly published (Struble and Herr much each of the two texts contributes to a better under-
mann 2009; Pardee 2009).31 Each is a mortuary monu- standing of the other.
ment with a table scene and a West Semitic inscription,
yet the difference in the state of preservation could not The Table Scene and the Symbols Above
be more marked, dictating the character of the work re-
The quality of the carving is low, and seems to have
31 Vocalization of ktmw following Younger 2009. been so originally, even before the damage to the stele.

Two probably beardless figures face one another over a

small cross-legged table laden with bread (figs. 12, 13)
(cf. Orthmann 1971: 369). The flaps of bread are too
worn to make out; considering their overall thickness,
there were probably three.
The right-hand person is seated on a tall, high-backed
chair with a cross-beam at mid-height of the legs. The
person’s feet rest on a footstool. Most details of the face,
hairstyle, and headdress are gone. All the same, the lack
of a veil is unmistakable. The figure wears an ankle-long
garment, its sleeves and collar or opening worn away.
The feet are visible, but have lost all detail; the footwear,
if it ever existed, is now indiscernible. In its right hand,
the figure seems to hold two lotus flowers (forming a
trio with the flower held by the person opposite). This
is in some doubt, however, and we have no alternative
to offer. The person definitely does not hold a cup. In its
left hand, the figure holds two narrow, straight objects,
their elongated tops thickened—spindles (Fig. 12a, b, c)
in view of several parallels (see comparanda in Figs. 14,
15),32 less likely ears of wheat. These objects obstruct
the right shoulder; as a result, the figure’s right forearm
oddly springs from the thigh.
The person on the left is standing. The outline of the
head is visible, but many details of the face, coiffure, and
headdress are obliterated. The figure wears an ankle-
length garment, held at the waist by a broad belt whose
end hangs down two-thirds of the way.33 The sleeves and
the collar or opening are worn away. The feet are visible
but have lost all detail; the footwear, if it ever existed, is
now indiscernible. In its right hand, the figure holds a
down-pointing mace with a long handle, its other end
protruding a little beyond the globular, exceptionally
large macehead.34 In its left hand, the figure holds a lo-
tus flower, as noted. The standing person fills the entire
32 Comparable spindles are to be found, for instance, in the hands

of the women on the orthostat from the outer citadel gate at Zincirli
(Orthmann 1971: Zincirli B/3; Fig. 14) and the orthostat confiscated
at Örtülü village but possibly stolen from nearby Zincirli (Kalaç 1975:
187–88; Fig. 15). Not poppy heads, suggested by Orthmann (1971:
375), as these capsules are given a circular shape when in relief. Also
on account of the mirror she holds in her right hand, the person on the
right in Fig. 15 is to be identified as a woman. Fig. 11 (a–c). Istanbul Archaeological Museums photos: (a) front,
33 The uninscribed colossal statue from Zincirli (Orthmann 1971:
whole: unnumbered.
Zincirli E/1; Bonatz 2000a: no. A 6) displays a comparable tassel,
though it is more elaborate.
34 We follow von Luschan (Sendschirli 4: 330), Lidzbarski (ESE 3:
194), and Orthmann (1971: 369) in identifying this object as a mace. It A down-pointing mace is the exception; even in the second and
is not a ladle, as suggested by Bonatz (2000a: 40), on account of a pre- first millennia, when the mace served almost exclusively as a staff or
sumed parallel from Carchemish (Orthmann 1971: Carchemish Ab/4), scepter (and as weapon of the gods), it habitually points up or is held
nor a “lute(?)” as hypothesized by Genge (1979: 101), though in n. 413 more or less horizontally. Outside the land of Sam’al, we know five de-
he quoted von Luschan’s and Orthmann’s proposed mace. The circular pictions of down-pointing maces—at nearby Sakçagözü (one), Kara-
Ördekburnu depiction is probably of a spherical macehead made of tepe (three), and Carchemish (one) only. Four of the parallels show
stone. The object is besides unduly large, but we have to remember that profiled heads of the type made of metal, while one of the Karatepe
the relief is also otherwise ill proportioned. examples is spherical like ours (Orthmann 1971: pls. 15h, 17g, 31e, 50a;

Fig. 11 (continued). (b) front, inscription: unnumbered.

Fig. 11 (continued). (c) front, inscription: Istan. 2072.

Fig. 12. Photographs of the scene: (a) 596; (b) 677; (c) 2920; (d) 677; (e) 2920 (photos by the authors).
Bossert 1942: no. 888; Çambel and Özyar 2003: 65, 77, 92, pls. 28–29,
frame of the relief, from top to bottom, whereas a space
52–53, 112–13). The raison d’être of this stance of the mace remains to may be discerned above and below the seated person.
be elucidated, however.
Down-pointing maces in the hands of a Sam’alian soldier (Orth-
mann 1971: pl. 60e), and an Assyrian soldier on the Lachish relief from We are grateful to Annie Caubet and Nicolas Grimal, whose
Nineveh, slab 9 (Uehlinger 2003: 284) are probably unrelated: Unlike pointed questions led us to investigate the distribution around Sam’al
the Ördekburnu mace and its Syro-Hittite counterparts, these maces of down-pointing maces, and to Tallay Ornan, who guided us to the
are held high. Lachish mace and Uehlinger’s paper.

Fig. 13. Drawing of the scene (by Rodica Penchas and Yulia Gottlieb Fig. 14. Orthostat from Zincirli, outer citadel gate. Istanbul Archaeo-
guided by the authors). logical Museums 7778 (Sendschirli 3: pl. 37c left).
The mortuary character of the stele was already rec-

ognized by Felix von Luschan, the excavator of Zincirli.35
The seated person is thought to depict the deceased, and
the standing person a surviving relative who performs
the rites (e.g., Bonatz 2000a: 40–41, 116–17). Among
Bonatz’s nine examples of steles with a seated deceased
facing a standing attendant not holding a flywhisk (Type
3b, Catalog C50–C58, pp. 40–41, 59; our stele is no. C52),
the majority of the seated persons were classified as ei-
ther female or of uncertain gender; only C50 definitely
depicts a male. If the objects held by the seated person in
the Ördekburnu stele are indeed spindles, they identify
her as a woman.36 As regards the standing person, a mace Fig. 15. Orthostat, from Zincirli? (confiscated at nearby Örtülü vil-
or scepter held by a mortal in ancient Near Eastern imag- lage). Gaziantep Museum 5600 (with permission, from Bonatz 2000a:
no. C26).
35 The stele is discussed with the “Grabstelen” in Sendschirli 4, and
the scene classified as a “Totenmahl” (see n. 2). Thus, see also Lands- ery is, with very few exceptions, a male marker.37 On the
berger 1948, cited in n. 27 herein (“Grabinschrift”). Lidzbarski, on the mace in that role, see Sebbane 2009: 304.
other hand (ESE 3: 197), hesitated between a mortuary and a votive
character. In the latter case, he considered the seated figure to be a god, mentioned by von Luschan (Sendschirli 4: 330), and paid no attention
possibly Rakibʾil, faced by a worshipper (ESE 3: 195), or else that the to the objects held in the left hand of the deceased.
two figures are a ruler and a high official (ESE 3: 195–96). 37 Despite the mace, von Luschan (Sendschirli 4: 330) considered
36 The lack of a cylindrical headdress convinced Lidzbarski (ESE the standing figure “perhaps female,” with no explanation. Orthmann
3: 195) that our seated person is a man—incorrectly, as it turns out; (1971: 375) remarked that the mace indicates the high rank of the man.
see discussion on the uninscribed Zincirli stele in the main text below, This is obvious: mortuary monuments will have been the privilege of
with n. 38. Genge too (1979: 101) was in favor of the identification as a the higher classes anyway, and the “servants” depicted in these monu-
male, referring to an alleged (but, in our judgment, nonexistent) beard ments are believed to be, in fact, family members.

Fig. 16. Mortuary stele from Zincirli, Hilani I. Berlin, Vorderasiatisches Museum VA 2995 (Sendschirli 4: pl. 54).
The uninscribed mortuary stele of a woman (Fig. 16) of the Ördekburnu deceased. The attributes mentioned
found close to a robbed-out tomb adjacent to Hilani I at in n. 38, combined with the first-rate workmanship and
Zincirli38 may have a bearing on the question of the status
made a convincing case for a woman despite the lack of a veil, mirror, or
38 Orthmann 1971: Zincirli B/3; Bonatz 2000a: no. C46. While spindles. See, in particular, Bonatz 2000a: 39 with n. 24, on the female
Orthmann (1971: 374–75) favored the identification as a male, Voos attributes of the seated figure that include the pleated dress and multi-
and Bonatz (1986 and 2000a, respectively, passim)—if hesitatingly— tiered bead necklace. See further Gilibert 2011: 93–94.

palatial find-spot, may well point to her identification Identification of the Individual Letters,
as royal, whereas the quality of the Ördekburnu relief is Tentative Drawing, and Transliteration
inferior, in particular the right forearm oddly springing
from the thigh. It should also be remembered that no The text on the front of the stele is much worn, as
mortuary stele showing a king at table has been securely noted (the right part more than the rest), and Lidzbar-
identified among the dozens of Syro-Hittite steles already ski’s observation about a measure of subjectivity in any
known, inscribed or uninscribed.39 decipherment soberingly true.42 Even much of line 10,
Three much-worn symbols, originally perhaps five or inscribed on the side of the monument and thus not nec-
more, are depicted in the top register (Figs. 12d–e, 13). essarily damaged by the felt-making, is no less blurred.
They could be divine symbols on comparison with the Working under better conditions than Lidzbarski’s (both
Zincirli reliefs of Kulamuwa and Barrakib (for which see in Istanbul and at home)—electric light, unlimited pho-
Yadin 1970; Tropper 1993: figs. 5, 16). If so, this is ei- tography—we hope to have counteracted the said hurdle
ther an indication that such symbols occur on non-royal more effectively.
monuments as well or, conversely, that our stele is royal; Lidzbarski’s three photographs (ESE 3: pls. 13–15),
there are at present no means for deciding. his drawing of line 10 (ESE 3: 194; our Fig. 8) and his
If their role as divine symbols is maintained, the transliteration (ESE 3: 197), Bossert’s partial trans-
middle one could be the yoke known from Kulamuwa’s literation and drawing of lines 5–8 (1932: 36, 87; our
and Barrakib’s monuments; this was already noted by Fig. 9), Lemaire’s 1980 notes and the five Istanbul Mu-
Yadin (1970: 210 with fig. 7) and before him by Lidzbar- seum photographs sent him in the 1980s (Figs. 1b1, 10)
ski (ESE 3: 196), who did not identify it, however. The constituted—besides the original—the starting point of
two flanking symbols, nearly obliterated, appear to us our work with the inscription. A new facsimile of all 10
circular and as such to portray celestial bodies—e.g., a lines appeared to us essential for conveying to the reader
crescent and disk or a star in a disk, well known in the the rationale of our proposed transliteration and transla-
royal Zincirli monuments. Lidzbarski, followed by Yadin, tion. The drawing was begun in Istanbul (Fig. 17), where
identified the left-hand symbol as a crescent and disk, we have sketched the front inscription on tracing paper
which we have been unable to see. The right-hand one placed over enlarged copies of Lidzbarski’s two close-up
was considered a plain solar disk by Lidzbarski, yet it photographs (ESE 3: pls. 14, 15). The clearest letters—
was restored as a god’s horned headgear by Yadin. In its about half of the total—were copied after repeated veri-
present worn state, it appears to be plain inside, so that fication against the original, illuminating each letter from
Yadin’s restoration is to be regarded as very uncertain.40 different angles. The side inscription we have examined
But was it plain originally, as Lidzbarski seems to have attentively, taking notes and sketching most letters, but
thought? This is not very likely: in stone monuments, we did not attempt a tracing.
the moon disk is often plain when resting in a crescent, The drawing of the inscription was improved and
whereas the other celestial bodies usually bear distinct completed from photographs, and the scene and out-
patterns inside. In theory, Yadin’s restoration of a winged line of the stele added, at Tel Aviv University by Rodica
disk above the yoke is not to be excluded, but on the stele Penchas and Yulia Gottlieb under our guidance. The
itself there is no trace supporting this. photographs were those just mentioned, as well as the
Bossert (1932: 35, n. 2) has considered the possibility ca. 200 images of the entire inscription or of details that
that the middle one of the three Ördekburnu symbols we took in Istanbul, illuminated from various angles. In
was a hieroglyph. Laroche (1960: xxviii) believed that all the following, these latter images are referred to by their
three were hieroglyphs, and Meriggi (1975: 220), while numbers only (automatically attributed by the cameras),
pondering the would-be hieroglyphic character of one without sigla (the photo numbers in bold refer to those
of the motifs, ultimately decided against it.41 We believe that are illustrated herein). The photos taken by Lidzbar-
that he was correct.
42 See n. 10. Moreover, near the beginning of line 2 there is a softer,
39 These steles stand in contrast to six gate-orthostats from Malatya, more porous streak in the basalt, running from top right to bottom
Carchemish, Zincirli, and Karatepe, where the same scene, presumed left, terminating in line 5. As a result, several letters have disappeared
royal, is found several times (Sass 2007: 308). The so-called Örtülü re- here, and some of those that remain display a false oblique appearance,
lief with a table scene (cf. n. 32), published in 1975 yet overlooked in which mirrors in fact the texture of the stone at this point. A similar
Sass 2007, should probably be added to the gate-orthostats. streak exists between lines 5 and 7. The two streaks are indicated in the
40 Meriggi (1975: pl. 39:287) has drawn the symbols, with the drawing (Fig. 18b, c). Some areas of the stone suffered damage prior
right-hand one somewhat resembling a divine cap. To us, this seems to the felt-making: The beginning of lines 1–2 is smoothed secondarily,
imaginary. obliterating the text there (Fig. 18a). The last four letters of line 2 and a
41 “[É] più probabile che si tratti di simboli divini e non di gero- little less of lines 3–4 are written within a depression, apparently indi-
glifici etei.” cating a correction (Fig. 18a).

Fig. 17. The authors at work in the Istanbul Archaeological Museums, April 2008 (photo by Evren Kıvançer).
ski and those supplied by the Istanbul Archaeology Mu- there is no guarantee that we have always been success-
seums are preceded by “Lidz.” and “Istan.,” respectively. ful. Indeed, it was only after thoroughly familiarizing
The illustrations of the inscription, six in all (Figs. 18, ourselves with the mostly blurred inscription that we felt
19), include a drawing, a photograph, and the former confident enough to decide about the shape of this letter
superimposed on the latter. But since no single photo- or that. At times, it seemed to us that a photograph il-
graph can bring out all the letters equally well, we cut luminated from a certain angle brought out the original
our many images into groups of a few letters each, then carving better, while another would rather accentuate the
traced each individual letter from the image or two that erosion of the stele at the same point (e.g., photographs
show it most clearly. This time (in Tel Aviv), we placed 689 and 671 for letter 7 in line 4).
the tracing paper over enlargements of two of our photos: While we hope that the two drawings of the entire
2901 of the front and 2911 of the side. Only the clearest inscription constitute a step forward, the uncertainties
letters, fewer than half, were drawn directly from these and alternative shapes cannot be expressed in them.
photos, the rest from the cut-out images (not reproduced When consulting Figs. 18c and 19c, one should there-
to scale here). Letters for which photos from different fore do so in conjunction with the letter-by-letter dis-
lighting angles render two or more shapes received as cussion and illustrations below. In the few cases where
many alternative drawings (e.g., letter 11 in line 1). A the two of us did not agree on a preferred identification
few of these are manifestly deceptive, and we have done of a certain letter, the said two figures reflect Lemaire’s
our best to distinguish the real from the imaginary,43 but view, with the alternative mentioned in the text be-
low and illustrated in the accompanying drawings and
photographs.
43 E.g., the arrow shape in line 6, position 2 (see n. 86). Also the The presentation of each line of the inscription in-
apparent circular depressions, around which the eye may reconstruct cludes a numbered transliteration, with Lidzbarski’s in
a circle, mistaking it for an ʿayin (see n. 106 on line 9, position 13). the first row, followed by ours. Reasonably clear letters are

Fig. 18 (a–c). Front inscription: (a) photo 2901.
transliterated in black type, uncertain ones in gray. Even • Position 2. The letter looks like a good nun in images
some of the former (black) letters may have more than 025, 026, 027, 649, but in others (e.g., 644, 645, 666,
one possible reading, depending on the lighting, e.g., 667), one hesitates between a broad-headed nun and
line 1, position 2. narrow-headed mem.44
Line 1 Positions 3–7. We estimate five obliterated letters be-

The beginning of lines 1 and 2 is worn smooth (photos tween the nun (or mem) in position 2 and the nun in
649, 2901, 2909). For the beginning of line 2, this was position 8, but the separation of these letters or even their
already recognized by Lidzbarski (ESE 3: 198). count has not been ascertained. At least one word divider,
now disappeared, is expected. Lidzbarski’s estimate (see
• Position 1. The traces visible now (photos 646, 647,
2901, 2902, 2910) do not correspond to the shape of
Lidzbarski’s proposed alep or to those of any other let- 44 In photos Istan. 346 and 2041, and our 035 and 2909, positions
ter, intact or damaged. Could it be not a letter at all but 1–2 may look together like a large mem with a vertical leg, certainly an
a delimitation of the inscription? optical illusion on comparison with the other photos.

Fig. 18 (continued). (b) drawing superimposed on photo.
chart above) was somewhat different. Here the name of Lidzbarski regarded the former as an alep or shin,
the deceased will have been written, the lineage, etc. the latter as a word divider.46 We too see the dot that
he has taken for a word divider, though on the stone
• Positions 3–5. Illegible. Judging by the joint horizontal and in most images it looks like the raised center of
measurement, we assume that there originally were a broad, low x-shape occupying positions 4 and 5 to-
three letters here. On the right (position 3), a shaft gether; this renders no legible letter, however.47
is visible, slightly leaning top right (e.g., photos 583,
644).45 As to the next two signs (positions 4 and 5), assume that the inscription begins with position 2 (see discussion of
position 1), a ṣade of “nṣb” might have been expected in position 3.
45 The sign was left unread in Lidzbarski’s transliteration, while in Alas, no wavy horizontal is visible and, in any case, the vertical is too
the text (ESE 3: 197–8) he remarked that a kap (because of the expected close to the preceding letter to make this possible. Similar difficulties
ʾnk), while improbable, was not impossible. We do not see how a kap arise for a purported mqm in positions 2–4.
can be reconstructed here. Photos 583 and 644 show this shaft with 46 In the transliteration, Lidzbarski marked positions 4 and 5 with
a dot attached to its right—hence a partly obliterated taw(?)—while dashes.

in others (clearest in 655), it may look like tiny he—firmly excluded 47 In most photos (e.g., 006, 583), the nearly vertical shaft of no.
by Lidzbarski (ESE 3: 197) and indeed an optical illusion. Should one 3, combined with the two right-hand arms of the x-shape, would look

Fig. 18 (continued). (c) drawing (by Rodica Penchas and Yulia Gottlieb, guided by the authors).
Positions 6–9. Lidzbarski saw a kap, and a word di- • Position 6. An obliterated letter. Certain photos (e.g.,
vider, and a nun, and a word divider.48 We assume four 022, 577, 655, 656) appear to display a shape resembling
letters here, of which only the nun in position 8 is certain, a right-angled waw; this is probably a phantom, however.
and we see no assured word dividers, even though one • Position 7. An obliterated letter (images 656, 674) in
(or two?) may originally have existed. place of Lidzbarski’s word divider.
• Position 8. The nun is one of only three or four clear
letters in line 1 (e.g., photos 655, 656, 668, 674), with
a break at mid-height. A circular hole is found in the
together like an oversized resh, whose left-hand tip merges with the stone immediately to the left (it has nothing to do with
hypothetical word divider. (These two right-hand arms are probably part the inscription, but it is helpful for orientation).
of Lidzbarski’s proposed alep.) In photos 577 and 674, part of positions
4–5 might look like a taw, the purported word divider constituting its
• Position 9. Much worn, either a small taw (photos 037,
center and right crossbar. The credibility of all these identifications is low. 685, 2908) or a divider (latter Lidzbarski; cf. photos
48 Lidzbarski had no comment in the text for the rest of line 1. 580, 653, 674).

Fig. 19. Side inscription: (a) photo 2911; (b) drawing superimposed on photo; (c) drawing (by Rodica Penchas and Yulia Gottlieb guided by the
authors).
Positions 9–11 are much damaged. Lidzbarski saw two 649, 667, 2901, 2920),49 less likely a strongly mutilated
rather than three characters after the nun (no. 8) and be- ḥet or nun (the latter two not drawn); then, possibly, a
fore the possible bet (no. 12). If indeed our position 9 word divider (e.g., photo 2908).
equals his word divider, we can match his bet with ours • Positions 14–15. Unidentifiable; indistinct remains
in position 12. only (e.g., photos 579, 643, 664, 2908). At position
14, the surface of the stele has partly flaked off, tak-
• Position 10. We have no reading for the mostly oblit- ing with it the bottom of the letter. This is probably
erated letter (e.g., photos 037, 650, 667, 2908). It what led Lidzbarski to identify a lamed or ʿayin here,
seems to be confined to the upper half of the line, i.e., two letters often confined to the top of the line. For
to possess no leg. position 15, he proposed a mem, which is probably
• Position 11. Only the lower part of the vertical shaft is not there.50
clear, visible in nearly all images. In photos 580, 647, • Position 16. A damaged letter with a vertical shaft,
649, and 653, we thought we could make out an alep, identified by Lidzbarski as a he, probably correctly
but this is very uncertain. In photos 643, 650, 671–73, (photos 651, 664, 665, 669, 670, 2908).51 The space be-
685, and 2908, this letter may look like a waw or kap tween positions 16 and 17 seems too small for another
(in 671 and 672 somewhat more like the former; in
643 and 2908 more like the latter), in photos 006 and 49 The worn letter in position 16 appears somewhat similar in cer-
674 like a nun. Of Lidzbarski’s suggested yod we see tain photos.
50 In photo 2908, one may see at position 14 the remains of a nearly
no trace. The space between positions 11 and 12 could effaced letter in the top two-thirds of the line, more or less circular in
accommodate a word divider; but if no. 11 is an alep shape, but this is an optical illusion: the bottom of this shape is none
or a kap, this would rather be its left tip. but the edge of the splinter that flaked away. Similarly, one may dis-
• Position 12. Possibly a bet, as identified by Lidzbarski, cern in position 15 what led Lidzbarski to identify a mem and why he
visible in most photos: a “leg” curved into an almost erred here: The practically effaced remains in mid-line could pass for
the oblique head of a mem, a feature that in nearly all other mems is
horizontal position fits the form of a bet, but the top of located farther up, and the ostensible oblique leg of the letter is again
the letter is broken away. For the state of the stone at an optical illusion: it is composed of the edge of the splinter that flaked
this point, see photos Istan. 2042 and our 670 and 674. away (top half) and a bump in the dividing bar between lines 1 and 2
• Position 13. A dalet or he. A letter with a vertical shaft (bottom half).
51 In several photos, the letter seems to have an oblique edge at
that Lidzbarski identified as a dalet or resh, of which
bottom left—oriented top left to bottom right—giving it the shape of a
the former is to be preferred (photos 650, 651), since dalet, but this is definitely due to a break (photos 577, 664; not drawn).
the top-left inclination does not favor a resh. Alterna- In some other images (e.g., 026, 035, 579, 649, 688, 2909), the head of
tively, a damaged he can be considered (photos 647, our letter looks tripartite in the same manner as does position 9 in line


‫י א‬ ‫ה‬ ‫ע מ‬/‫ל‬ ‫ר‬/‫ב ד‬ ‫– י‬ • ‫נ‬ • ‫• כ‬/– ‫ש‬/‫א‬ ‫כ‬ ‫נ‬ ‫א‬

‫י ו ב‬ ‫ה‬ – – • ‫ה‬/‫ב ד‬ • ‫נ‬/‫כ‬/‫ו‬/‫• – א‬/‫ת‬ ‫נ‬ – – – – – ‫נ‬/‫מ‬ –
19 18 17 16 15 14 ? 13 12 ? 11 10 9 8 7 6 5 4 3 2 1
letter; the tiny upside-down v-shape in photos 578, and probably misleading. This appears differently in
582, 642, 643, 651, 664, 2908, etc., is to us inexplicable. photos 582, 642, 643, 651, 664, and 2908, where the
• Position 17. The yod is clear in photos 578, 642, 643, bottom horizontal is invisible, or less visible, lending
651, 664, 665, and 2908. the letter the shape of a waw,52 but this is nearly as
• Position 18. Uncertain. We see the remnants that led doubtful.
Lidzbarski to identify an alep (photos 005, 006, 024, • Position 19? Yet another letter, a bet(?), may have been
2902, 2903), but they would be similar in form to squeezed in at the very end of the line (photos 674,
the alep in the Ahirom inscription, which is unlikely 2903, 2910), like the possible bet at the end of line 4;
or is this illusory?
10 (identified as a mem), but the significance of this, if any, remains
obscure. 52 Or a ṣade(?)—not drawn.

Line 2 • Position 6. The nun is certain, even though its top

• Position 1. An obliterated letter? As in line 1, the sur- stroke is not preserved in its entirety (Lidz. pl. 14 and
face is worn smooth here (clear in photo 2909, invis- photos 644, 648, 652, 666, 671, 684).
ible in 2910): “Vor ‫[ מת‬i.e., our positions 2–3] ist ein • Position 7. A small, much obliterated letter high on the
Stück weggeschliffen, da hat vielleicht noch ein Buch- line, probably a yod, as observed by Lidzbarski (Lidz.
stabe gestanden” (ESE 3: 198).53 pl. 14 and photos 037, 576, 644, 648, 666). The word
• Position 2. Unidentified. The letter, read mem by Lidz divider to the left is discernible in most photos; as be-
barski, seems to have a shin-like head and a curved tween positions 4 and 5, its ostensible extension to the
leg (photos 006, 553, 644, 645, 649, 654), the latter left is rather part of the following letter.
more suitable for a bet. But this form would be close • Position 8. A bet. Lidzbarski remarked on this let-
to the Mesha mem—out of the question here—and it ter (ESE 3: 198): “am ehesten ein ‫ב‬, weniger wahr
differs from the clear Ördekburnu mems. Photos 655 scheinlich ein ‫ט‬, noch weniger ein ‫מ‬.” We do not
and 2909 show best how very worn the surface here doubt the bet—see Lidz. pl. 14 and photos 037, 648,
is, so that much of the various lines seen in the other 650, and 684 (on the improbable ṭet, cf. n. 54). The
photos may be illusory. seeming dot visible between positions 8 and 9 (same
photos) is better understood as the tip of the lamed
Positions 3–4. Lidzbarski proposed to identify taw-yod in position 9.
here, two narrow letters. To us, this spot on the stone • Position 9. The lamed is reasonably clear in most pho-
is illegible, but the remains may indeed justify counting tos (e.g., Lidz. pl. 14 and 580, 648, 671, 673, 684).
two narrow letters as in positions 6–7, perhaps followed Lidzbarski’s second choice, a taw, was founded on the
by a word divider, rather than a sole wide one. continuation of the stroke downward, part of which
is indeed there (same photos) though probably unre-
• Position 3. Unidentified. The oblique lines, top right to lated to the letter (not drawn). “Hinter dem folgenden
bottom left, reflect the porous inside of the weathered Zeichen [i.e., position 9] steht sicher kein Punkt” (ESE
stone (photos 655, 2903, 2910), in a patch continuing 3: 198); we agree.
to position 4. One can see the traces that led Lidzbar- • Position 10. The ḥet is inscribed in a depression (e.g.,
ski to read taw, but it remains uncertain whether they Lidz. pl. 14 and photos 580, 647, 671, 673, 688). There
are manmade. may be room for a word divider thereafter, but no
• Position 4. Unidentified. As noted, we consider that trace of it remains.
the porous area continues to here: the inscribed sur-
face is completely effaced, exposing the texture of the Positions 11–13. There follow three unidentifiable let-
basalt underneath (the erosion is best seen in photo ters. This particular area of the stele is much damaged,
027). Touching the following bet, a word divider is and a reading practically impossible.
possibly discernible (photo 648). Near here, Lidzbar-
ski (ESE 3: 198) saw “vielleicht ein tief stehendes Jod,” • Position 11. Unidentified. Of Lidzbarski’s shin, we see
which we were unable to make out. its purported right-hand v-shape touching the top of
the line (e.g., photo 650). As there seem to be nearly
Positions 5–10. We largely follow Lidzbarski on the effaced traces of a letter below and to the left, we doubt
next six letters and word divider, bny•blḥ (Lidz. pls. that the v-shape belongs to them,55 but have no alter-
14, 15). native to offer.56
• Position 12. Unidentified. In Lidz. pl. 15, photo 2903,
• Position 5. The bet is damaged but unmistakable; see and several others, one can see the traces Lidzbarski
Lidz. pl. 14 and photos 644, 648, 667.54 considered an alep; the “chevron” is indeed there, but
what he has taken for the bottom end of the alep’s
53 In photos 2900, 2901, 2902, 2903, and 2910, two short horizon- 55 In certain photos (e.g., 671–73) the v-shape looks as if followed
tals may be seen, suiting remnants of a samek more than of a zayin, but by a large bet or dalet, but this is probably deceptive. Indeed, in Istanbul
they are low on the line and invisible in most other photos. we thought we saw waw and mem in positions 11–12, but this was so
54 Lidzbarski wavered between bet and ṭet in his transliteration (no uncertain that we did not draw them. Back home, we realized that three
comment in the text); likewise in the case of position 8. He evidently v-shapes just below the top line separator (photos 650, 671) seemed to
imagined the hypothetical ṭet as a stroke within a circle, whereas in us to constitute the heads of these letters.
our inscription one would rather expect a cross within a circle (see 56 In theory, there could have been two narrow letters here rather
“Palaeography” with Fig. 21). than a single broad one, or a single letter and a word divider.


v ertical is in fact sunken rather than in relief, invali- the remains of a nearly obliterated letter. A word di-
dating the suggestion. vider, on the other hand, has the benefit of supporting
• Position 13. Unidentified. According to Lidzbarski, a plausible reading of the following string if that string
“vielleicht kein ‫ת‬, sondern ein Punkt.” All we see is a begins with a waw (see next).
short segment of an oblique line, top right to bottom
left (photos 648, 652, 654, 665, 666, 684),57 possibly
57 In Lidz. pl. 15, a z-shape may be made out here, yet this is an sunk, not raised. Besides, all identified zayins in our inscription are of
optical illusion, for in our photos (e.g., 650) the said shape is clearly the archaic, sideways-H shape.

‫ב ח ל ב ב ה‬ ‫נ‬ •/‫ת‬ ‫א‬ ‫ת ח ש‬/‫מ ל‬/‫ט‬/‫• ב‬ ‫נ י‬ ‫ט‬/‫ב‬ ‫י‬ ‫ת‬ ‫מ‬ ?

‫ב ח ל ב ב ה‬ ‫נ‬/‫ו‬ •/– – – ‫ת ח‬/‫ל‬ ‫• ב‬ ‫נ י‬ ‫ב‬ • – – – –
20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 ? 4 3 2 1
• Position 14. A waw is a distinct possibility (photo Istan. ignoring the tiny vertical stroke instead. The best photo
no number,58 and our 671). If one accepts the rather is Lidz. pl. 15, showing the right-hand side of the bet
hypothetical divider to the right, this would yield a practically intact (which might indicate that the stone
plausible reading, wbḥlbbh.59 This is underpinned fur- has deteriorated somewhat in the past century).
ther by the lack of word dividers after positions 14 and
15.60 Lidzbarski (ESE 3: 192) concluded that “[e]ine Positions 16–20. The clear reading ḥlbbh, recognized
genaue Durchprüfung des Textes ergab das überra- by Lidzbarski, is clinched by parallels in the Hadad statue
schende Ergebnis, daß auf der Vorderseite kein einziges (ESE 3: 197, 201; Tropper 1993: 269). The last four letters
Wort mit Waw beginnt.” Indeed, the lack of waw was are written in a depression (see n. 42), which may also be
one of the purported characteristics that led him to de- the reason for the too-large distance between the ḥet and
cide about the largely non-Semitic classification of the the following lamed.
language (ESE 3: 192). We suggest three more potential
conjunction waws in lines 6, 8, and 9. • Position 16. The ḥet is clearest in photos 024, 553, 654,
• Position 15. For the bet, see photos Lidz. pl. 15, Istan. 669, 674, and 2910.
2041, and our 655, 671, and 684. Not only has the • Positions 17, 18. The lamed and first bet are clear in
letter lost much of its right-hand line, but a vertical most photos (e.g., 642, 650).
line seems to join the head of the letter to the bottom • Position 19. A bet. Clearest in photos 2900, 2902, and
stroke. All this could make the letter look like a zayin, 2903.
depending on which lines one attributes to it, or as a • Position 20. A he. Rather worn but unmistakable in
mixture of both letters (photo 674 and several others, photos Lidz. pl. 15, Istan. no number, and our 576,
e.g., 650, 651). We prefer the reading bet: the head of 642, 2909, and 2920. Photo 577 shows the letter shape
this letter cannot be overlooked, which leads in turn to differently.
58 Two among the five Istanbul Museum photographs of the front Line 3
inscription bear no number.
59 This is why we think Lidzbarski’s identification of a nun less • Positions 1–4. Of the four characters that Lidzbarski
likely, even though, as far as form alone is concerned, it could have made out before his first word divider, we can identify
been an option (e.g., photos Istan. 2041 and our 642, 643, 649, 688). In none, nor the divider.61 We cannot even tell whether
photos Istan. no number, and our 671 and 674, the letter may look like
a cross between waw and nun. 61 In Lidz. pl. 14, the “ghost” of a waw could be made out in po-
60 Lidzbarski too (ESE 3: 198) remarked “zwischen ‫ נ‬und ‫[ ב‬our sition 4 (not drawn); compared with the other photos, this shape is
positions 14 and 15] sicher kein Punkt.” probably illusory.


his count was right: there may have been four letters short leg could favor a dalet. Lidzbarski’s yod, on
without a divider or three and a divider. For the state the other hand, is doubtful,62 but it fits the reading
of the stone, see photo 576. ysʿd in p
ositions 5–8, a form already documented in
• Position 5. Either a much worn dalet (photos 579, 647,
62 The traces that made Lidzbarski read a yod can be discerned in
653), not unlike no. 15, or a he (e.g., photos Istan. 2041
Lidz. pl. 14 and photo 2900. But the “back” of the presumed letter (the
and our 684). Most of the letter is inscribed within long oblique, top right to bottom left) is, in fact, the groove marking the
the damaged zone—hence the uncertainty; only the right-hand edge of the damaged zone in the stone. This groove bisects
bottom of its leg is located without. The seemingly the leg of the original letter, which is inclined the other way.

‫ה‬ ‫י‬ ‫ר מ ל‬ ‫ז‬ ‫ל‬ • ‫י‬/‫פ ה‬ ‫ר • ה כ‬/‫ד‬ ‫ש ע‬/‫ס‬/‫מ‬ ‫• י‬ ‫מ – ש‬ ‫א‬

‫ה‬ ‫י‬ • ‫נ נ ל‬ ‫ז‬ • ‫ל‬ ‫•ע‬/•‫נ פ י‬/‫כ‬/‫ד • ה ו‬ ‫ע‬ ‫י ס‬/‫ה‬/‫ד‬ – – – –
19 18 ? 17 16 15 14 ? 13 ? 12 11 10 9 8 7 6 5 4 3 2 1
Sam’alian.63 A word divider may or may not have fol- Positions 7–8 were read ʿd by Lidzbarski, and he was
lowed. probably right, certainly for the ʿayin.
• Position 6. Probably a samek. Lidzbarski remarked that
“[v]or ‫[ עד‬our positions 7–8] einerseits Züge wie von • Position 7. The ʿayin looks its best in photos 603, 646
einem [Schin], andererseits scharf ein kurzer Strich and 2901; it appears rather poor and squarish in pho-
darunter. Das Bild [Schin] ist vielleicht durch spätere tos 656 and 671.
Vertiefungen im Kopfe eines ‫ ס‬entstanden” (ESE 3: • Position 8. Probably a dalet. “‫ ד‬mit kurzem fuß, doch
198). Or else—according to the transliteration and kann freilich ein Stück abgesprungen sein [i.e., origi-
the mention of a “Punkt vor ‫[ ימ‬our positions 5–6]”— nally a resh]” (ESE 3: 198). To us, the letter is too much
Lidzbarski may also have considered a mem. In Lidz. inclined top left for a resh; it may look like a dalet,
pl. 14, we see what motivated Lidzbarski to mention rather unsuccessful and a little too large, in photos
the shin, but then doubt it and propose mem (not un- 036, 037, 667, 671, 684, 685, and 2901.65 The letter is
like position 2 in line 2). In the same photo, we also followed by a word divider, observed by Lidzbarski,
see what led him to the alternative samek, and this clearly figuring in photos 671 (illustrating letter 7)
appears to us to best fit the remains (cf. photos Istan. and 684, among others.
2042 and our 579, 603, 2909).64 The right-hand 20% • Position 9. A he; see photos 671 (illustrating letter 7)
or so of the head of the samek, located in the damaged and 672, among many others. Of its vertical, the top
zone, has disappeared. two-thirds (or three-quarters) have been knocked
away.
• Position 10. Either a waw, kap, or nun, none particu-
63 While we doubt the yod, sʿd in positions 6–8 remains our favored
larly appealing; see photos 583, 671, and 672—with
reading in any case.
64 In some of the photos, the letter looks rather like a he (e.g., Lidz.
pl. 15 and photos 580, 646, 655), yet with its leg inclined top right (i.e., 65 In photos 2909 (illustrating letter 6) and Istan. 2041, it seems to
contrary to the norm). This false impression is created by the sharp possess a lower horizontal and thus looks like a bet. In yet other photos
oblique groove immediately to the right that has obliterated this side (003, 688), the head appears to be open on the left, lending the letter
of the letter. It marks the left-hand edge of the porous zone, already the shape of a legless he, or mirror E. All this is probably deceptive,
mentioned, in the stone. however.

the slightly oblique shaft, there are three possibilities of the letter creates the illusion of a closed head and,
for the head.66 what is more, the top-left vertical of the nun is hardly
• Position 11. Uncertain. With much good will, one apparent.
might make out the head of Lidzbarski’s pe (Lidz. pl. • Position 16. Probaly a nun. The letter is best visible in
15 and photos 035, 655, 2900),67 but this is far from photos Istan. no number and our 578, 643, 665, 684,
ascertained. The hopelessly damaged surface is visible 687, 2908, and 2920, where nun seems clearly prefer-
in photo 576. able to Lidzbarski’s mem.69 The head of the letter is
• Position 12. Two very different possibilities, yod- fairly broad but sketched as that of a nun. The reading
divider or divider-ʿayin, demonstrate again how un- is not absolutely certain, however: in photos 649 and
certain much of the reading is. What may look like a 2901, it looks less like a nun, though not more like a
yod is barely discernible in Lidzbarski pls. 14 and 15, mem, and we have no alternative to offer.
and in our photos 579, 656 and 671–73. It is possibly • Position 17. A lamed (photos 2900, 2901). Followed by
followed by a vertical word divider (photos 656, 671– a tiny dot (also photo 2908), so inconspicuous that it
73), as already observed by Lidzbarski. But neither the may be accidental.
letter nor the divider is certain, and in other photos • Position 18. A yod (photos 649, 2900).
(e.g., 576, 579, 666, 2901), the same traces may look • Position 19. A he (photos 649, 666).
like an ʿayin, a word divider to its right, with the ensu- • Beyond position 19. Left blank originally (photos
ing reading ʿl znn in positions 12–16 (though weak- Istan. no number and our 2901). Thus, already Lidz-
ened by the lack of a word divider at the end). The left barski (ESE 3: 198): “[H]inter ‫ ליה‬kein Punkt.”
part of this ʿayin is the vertical stroke just mentioned
as a possible word divider following the purported Line 4
yod, in different lighting. But the ensuing shape may • Position 1. Probably a pe (photos 576, 669). In the
in fact be another of the several cases where the eye same photos and in 2903, with a seeming additional
connects disparate features around a depression in the stroke at the top, the letter may look not unlike a nun,
stone, with misleading results (see below on line 9, Lidzbarski’s second choice.70 The rounded right-hand
position 13, with n. 106). “shoulder” could suit one of the variants of pe better
(an angular variant exists), yet two other nuns, posi-
Positions 13–19: lznnlyh? The zayin and yh (14, 18– tion 3 in line 5 and position 2 in line 8, are similarly
19) are among the clearest letters in the inscription, the rounded. The nun may have an advantage: with an-
two lameds (13, 17) feeble but unmistakable, letters 15 other rather hypothetical identification in position 5,
and 16 possibly readable as two nuns whereas Lidzbar- the end of line 3 and beginning of line 4 could read
ski proposed resh-mem. This sequence is certainly in- lyhny ʾš[m].
terspersed by a word divider or two, but their location • Position 2. A yod (photos 576, 669, 2903).
is debatable. • Position 3. An alep (photos 576, 580, 690).
• Position 13. A lamed (photos 666, 2901). On the word Positions 1–3: Word dividers? The oblique bottom
divider, see next. stroke of the yod (position 2) is quite long, so that what
• Position 14. A zayin (photos 649, 2908). The top hori- appears to be its right-hand tip may just have merged
zontal, to the right of a break, seems to extend more with a worn-down word divider. The bottom line of the
to the right than usual, or is it (the “tip”) in fact a word alep’s chevron merges with the top horizontal of the
divider? A divider at this point would fit the proposed, yod—again, unless a word divider, now mostly worn
yet uncertain, reading ʿl znn. smooth, separated the two originally. Between the alep
• Position 15. Almost certainly a nun (clearest in photo in position 3 and the shin in position 4, there is also room
Istan. no number). Several photos (Lidz. pl. 14, for a word divider, though the dot that we have drawn
578, 649) make it plain why Lidzbarski opted for a seems to be located too high. Apparently, only one of
resh here:68 A shadow within the right-hand angle these potential dividers is real; in “Tentative Translation
. . .” we opt for the second.
66 For waw, an oblique shaft would be less common than a strictly
vertical one, but still admissible.
67 Hypothetically, though probably less likely, the same strokes may
be what remains of a yod (not drawn). 69 Lidzbarski has not commented on this letter in his text.
68 In the transliteration only; Lidzbarski has not commented about 70 “Der erste Buchstabe weniger wahrscheinlich ein ‫( ”נ‬ESE 3: 198).
this letter in his text. In any event, we disagree with Lidzbarski’s first choice of a taw.



‫י‬ ‫נ ב כ‬ • ‫ב‬ ‫י‬ – – ‫י‬ ‫ש ח‬ ‫י • מ‬/– ‫ש‬ ‫א‬ ‫נ • י‬/‫ת‬

‫י‬/‫ב‬ • ‫נ‬/‫נ ב כ‬/‫ו‬ • ‫י ר ב‬/‫ס – • – ב ח‬ – – ‫ש‬ • ‫• א‬ ‫נ • י‬/‫פ‬
17 ? 16 15 14 ? 13 12 11 10 9 ? 8 7 6 5 4 ? 3 ? 2 ? 1

• Position 4. A shin (e.g., photo 690). • Position 7. A samek? This could be, if one imagines
• Position 5. Unidentified. “Hinter ‫[ יאש‬i.e., positions the letter to possess a bottom vertical (photos Istan.
2–4] vielleicht ein Jod,”71 indeed a possibility (Lidz. 346 and 2041, and our 011, 028, 035, 579, 580, 585,
pls. 14, 15 and photos 008, 037, 684, 685, 690), if a 689, 696, and 2909). The said feature may be only
remote one, for the letter seems inclined the wrong the shadow of a crack farther left that continues also
way. Or perhaps it is (see photos 584, 585, 647, 671, downward below the line separator, but with the state
673, 689, 693, 2909) another letter altogether, one of the stone one can always argue that the leg was bro-
that may recall a waw. In photos 647, 689, and 2909, ken away. The samek we have drawn is thus far from
one of us (Lemaire) sought to identify a mem, or else certain, yet not impossible. In different lighting, the
a resh (neither is drawn),72 for the sake of a tempt- leg nearly disappears, and a notch becomes visible at
ing, if speculative reading mentioned with position bottom center of the body of the letter, with an en-
1. Yet, in fact, all these are patchy remains (best suing small ḥet (photos 006, 008, 029, 037, 552, 559,
seen in photo 2909 illustrating letter 6), which the 577, 581, 583, 584, 644, 645, 648, 652, 654, 666, 668,
eye may connect in different ways, each probably far 690, 2920).73 Between samek and ḥet, we prefer the
from reality. former: The photos exhibiting it seem to us to echo the
original engraving; the photos that bring out the ḥet
Positions 2–5. “Ich habe die Stelle auf ‫ באשר‬geprüft shape seem rather to reflect the subsequent damage.
[based on the assured alep-shin sequence in positions Photos 009, 663, and 671 show best the state of the
3–4], doch scheint mir diese Lesung ausgeschlossen” stone at this spot.
(ESE 3: 198). Excluded indeed: While the alep and shin • Position 8. Much mutilated, unidentified. In several
are certain and the resh a faint possibility, the letter in photos (e.g., 028, 029, 035), the letter seems to con-
position 2 is a practically secure yod, and bet is out of sist of two or three superimposed horizontals, a can-
the question. Lidzbarski had no comment on the rest of didate for a samek (hence also Lidzbarski’s ḥet), quite
the line. similar to the preceding letter. The same photos may
Positions 6–8 (Lidzbarski’s mšḥ) are to us barely show the letter followed by a word divider, or else
legible: that spot may be part, now damaged, of the next let-
ter. For the state of the stone, see photos 009, 576,
• Position 6. Illegible. The traces of a long, oblique leg 652, and 2900.74
(photos 552, 553, 581, 668, 690, 2904) may be identi- • Position 9. Unidentified. Photos 011, 028, 035, 579,
fied with various letters, including a mem as suggested 585, 689, 691, 696, and 2909, among others, ostensi-
by Lidzbarski, but these readings remain conjectural: bly display Lidzbarski’s yod, but photos 655 and 656
the left side of the letter is completely effaced, located seem to show a pe, and photo 663 and others show
as it is in the porous streak running from line 1 to an unhappy union of both. Neither is particularly
line 5. Alternatively, in photos 553 and 668, admit- convincing.
ting other traces, one may perceive a yod. In photos • Position 10. A bet seems quite discernible in photos
Istan. 2041 and our 003, 579, 696, 2909, and 2920, a 011, 579, 580, and 585, among others. Lidzbarski left
large circle is visible there that may be mistaken for an the letter unidentified.
ʿayin (cf. n. 106), and in photos 559, 577, 685 (illus- • Position 11. Unidentified (thus already Lidzbarski).
trating letter 5), 692, and 2908, it seems to have a cross Different lighting may accentuate different shapes,75
shape inside, an imaginary ṭet. In photos 554 and 575, but in fact, the specific place on the stone is practi-
the same may give the impression of an oversized bet. cally effaced.
Still different, unidentifiable shapes, which we have
not drawn, are to be seen in photos Istan. 346 and our
009, 671, 672, 673, 693, and 2907. This wide spectrum
73 Lidzbarski considered it a small shin, significantly smaller than
of possibilities indicates that the stone is hopelessly
the letter in position 4.
damaged here. To the left, inside the softer streak in 74 Photos 652 and 2900 show an ostensible narrow nun in the right-
the stele, there may be room for a word divider, but hand half of the space occupied by the letter; to us, this appears to be
nothing can be seen. part of the erosion.
75 In photos 010 and 2902–3 a ḥet may be inferred, in photo 645
the remains of a yod; both are uncertain. The reading with a ḥet yields
a tempting bḥrb, however. In Lidz. pl. 14 and photos 003, 553, and 663,
71 InLidzbarski’s transliteration, merely a dash. one can clearly see a circular depression in mid-line, and with some
72 A resh was Lidzbarski’s initial instinct, subsequently discarded good will restore an ʿayin around. But the depression is likelier to be
(see next paragraph in main text). the result of a blow, one of many, to the stone (cf. n. 106).

• Position 12. A resh. Quite clear in photos 003, 033, others (e.g., 011, 028, 033) the spot preceding posi-
034, and 649, the head also in photo 2907, and the leg tion 2 looks hopelessly blurred.
is especially clear in 696 and 697.76 • Position 2. A he, partly broken but unmistakable. The
• Position 13. The bet is clear (e.g., photos 003 and 033). best view is in photos 011, 028, and 035.
There is ample room to the left for the word divider • Position 3. Probably a pe or nun (photos 011, 635,
Lidzbarski has indicated, but its location remains 640, nun clearest in the former).80 In the transliter-
questionable (with more than one possibility in Lidz. ation, Lidzbarski wrote kap and mem. In the text, he
pl. 15 and photo 003). remarked that “Der dritte Buchstabe könnte an sich
• Position 14. A nun (thus Lidzbarski) is probably trace- ein ‫ פ‬mit einem Punkte links oben sein, aber dann
able (Lidz. pl. 14, Lidz. pl. 15, and photos 009, 651, wiche die Form stark von der des folgenden Zeichens
2907, 2908), or else a waw (not drawn).77 ab, das doch nur ein ‫ פ‬sein kann” (ESE 3: 198). The
• Position 15. The bet is clear (e.g., photos 2901, 2902). shape that looked like a kap or mem to Lidzbarski (as
• Position 16. Identified by Lidzbarski as a kap, this letter an alternative to pe) seems to us rather to resemble
may alternatively be a nun followed by a word divider a nun. And regarding the pe alternative, we do not
(the latter instead of the kap’s left-hand “finger”),78 in share Lidzbarski’s objection to contiguous variants of
photos Lidz. pl. 15, Istan. no number, and our 005, the same letter (compare kbb in line 8). We see a simi-
034, 559, 577, 578, 649, 690, 2904, and 2905.79 Yet if lar hesitation between pe and nun in line 4, position 1.
the next letter is a yod, the said divider may alterna- • Position 4. Possibly a pe (photos 635, 640), with kap an
tively be understood as the right-hand tip of the yod’s alternative in photos 033 and 688 (cf. line 6, letter 12).
lower horizontal (photo 024). That the leg of the letter looks curved in some photos
• Position 17. Possibly a bet (e.g., photos 559, 575, 693, is due to a natural groove in the stone immediately to
2905) or a yod (Lidzbarski’s reading; photos 024, 553, the right; the leg is hardly bent at all, as best seen in
654, and 2904 show the head of the letter as if open photos 635 and 640. The letter is probably followed by
at left)—smallish, angular, and squeezed in high on a word divider (photos 033, 688).
the line. Less likely a resh, whose leg has all but dis-
appeared, or a triangular ʿayin if the bottom part of Positions 5–6. The two letters are only partly pre-
the letter is ignored altogether (the latter two are not served, but the reading hʾ, followed by a word divider, is
drawn). as good as certain on comparison with the parallel in line
6 (already Landsberger 1948; see our n. 27).
Lidzbarski (ESE 3: 200) proposed to read nbkyn in line
4, positions 14–17, and line 5, position 1. • Position 5. A he. The vertical is clear in photos 028 and
689; the middle horizontal is unusually thin.
Line 5 • Position 6. An alep, best visible in photo 635. The cen-
• Position 1. Possibly a nun. The leg is close to the stele’s ter of the letter is broken away (photo 2901). A word
edge, and the letter is only partly visible in most pho- divider follows (photo 635).
tographs. The whole is possibly visible in photo 640, • Most of the next letters, positions 7–15, are clear and
partly in Lidz. pl. 15 and photos 011, 635 (head), and their reading lrkbʾl•ʾlh is secure.
028 and 689 (leg). Alternatively a waw(?), if the top • Position 7. A lamed. The lower stroke is partly dam-
of the right-hand vertical, apparently visible in photo aged (photos 635, 655), but enough of the letter is
640, is indeed there, but we doubt it. A word divider preserved for identification.
seems to follow in Lidz. pl. 15 and photo 640, but in • Position 8. A resh with an unusually curved leg (pho-
tos 635, 655).
• Position 9. The kap is best seen in photos Lidz. pl. 15,
Istan. 2041, and our 635, 636, 644, 646, and 655.81
76 Lidzbarski identified the letter as a yod on the strength of the
• Position 10. A bet (photos 578, 663, 664, 665).
seemingly open head, whose opposite end appears besides to extend
much to the right (e.g., Lidz. pls. 14, 15 and photos 554, 655, 656). Both
• Position 11. An alep (photos 649, 663, 664, 684, 2907).
these traits seem to us deceptive. • Position 12. A lamed (photos 578 [illustrating letter
77 In photos 576, 690, and 2902, the letter seems to be too close to 9], 663, 671, 684, 2907). A word divider will have
the following bet, as if there were room for only part of it, but this is
probably deceptive. 80 What may look like a left-pointing trident shape at the top (pho-
78 Somewhat similar to letter 13 in line 9. tos 033, 688) does not belong to the original letter; it is probably the
79 Illuminated differently in photos 037, 674, and 2904, the same result of subsequent damage.
remains may resemble a taw, but this is unlikely, as the letter would 81 In photos 004, 578, 584, 663, and 664, the letter may appear as a
then be inclined top left instead of the expected other way around. circle, quite a common mirage ʿayin in our stele (see n. 106).



‫ה א • ל ר כ ב א ל • א ל ה‬ ‫פ פ‬/‫מ‬/‫כ‬ ‫ב‬/‫• ה‬ ‫נ‬

‫פ • ה א • ל ר כ ב א ל • א ל ה‬/‫כ‬ ‫פ‬/‫נ‬ ‫נ • ה‬/‫ו‬
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
followed in the now nearly obliterated spot (photos the photographs show the squeezing better than does
553, 2910). the drawing.
• Position 13. An alep (photos 553, 581, 583, 663, 672, • Positions 2–3. Possibly nun-ʿayin, or else a mem(?).
684, 2901, 2910). The shaft at right is clear; for the rest there are two
• Position 14. A lamed, either with exceptionally short possibilities: (1) A shaft and a single stroke yield
arms, or damaged (photos 663, 672, 673, 684, 2901, a nun—Lidzbarski’s preference, with the next two
2910). strokes combined, viewed differently, transformed
• Position 15. The identification of the letter as a he is into a mutilated ʿayin (Istan. no number and our 640,
clear, but it is partly obliterated. Its leg is curved bot- 690, 695, 2905, 2906).84 (2) Just a mem whose head
tom left (photos 553, 581, 2901, 2910).82 Situated at contains an extra stroke, a rare but not unheard-of
the end of the line, ʾlh is not followed by a word di- feature (e.g., WSS 1091) (same images),85 in which
vider, as already noticed by Lidzbarski (ESE 3: 198)— case, position 3 disappears altogether.86 Our first
see the end of the discussion of line 7. choice is nun-ʿayin, which yield a plausible reading.
• Positions 4–5. The former letter could be a mem or
Line 6 nun, the latter a waw or qop—six tentative readings
Positions 1–4. There have been varied identifications in all for the same drawing and photos 637, 638, and
of the letters and word dividers to the right of the op- 639: mem-waw, mem-divider-waw, mem-qop, nun-
tional divider preceding position 5, the main hesitation divider-waw, nun-divider-qop, nun-qop. The three
concerning position 3—a letter, a word divider, or zero.83 short strokes in the center of positions 4–5 may be
Three of the identifications yield readings: either nʿm for assigned variously: (1) The shaft and the rightmost
positions 2–4 (or ynʿm for positions 1–4), or ymm or of the three following strokes could constitute a nun,
ymn in case position 3 does not in fact exist. This multi- the middle stroke a word divider, and the left-hand
tude calls for caution. stroke may be part of the letter in position 5, in that
case a qop. (2a–d) The shaft and the first two strokes
• Position 1. Probably a yod (photos 637, 638, 639, 670,
84 Lidzbarski’s transliteration of line 6 begins with yod-nun-divider.
695, 2905, 2906). It is sculpted at the very edge of
In his text, he had no comment on these three signs. He has probably
the stone, its bottom horizontal barely visible. This identified our optional ʿayin in position 3 as a word divider, despite its
crammed position (expected at the ends of lines rather far too large size.
than at their beginnings) is repeated in line 8, where 85 The theoretical option of a habitual mem and a word divider does
not agree with the divider that we prefer to identify following the next
letter.
86 Photos 637, 638, and 639 create a wholly different impression: In
82 In photo 2910, the leg appears more curved than in photo 553. them, the shaft looks as if surmounted by a Λ-shape, which lends it the
83 Indeed, Lidzbarski counted our positions 1–4 as three letters and form of an upward-pointing arrow, perfectly clear in the photographs
a word divider—see n. 84. yet obviously deceptive.


‫ר מ ל י‬ ‫ל כ ב ב • א‬ • ‫• ה א‬ ‫• ט מ י‬ ‫כ‬/‫ו‬ ‫מ‬ • ‫י נ‬
‫ר מ ל י‬ ‫ל כ ב ב • א‬ • ‫• ה א‬ ‫נ י‬ ‫ח‬ ‫ק‬/‫נ • ו‬/‫• מ‬/‫נ ע‬/‫י מ‬
19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 ? 4 ?3 2 1

could constitute either a narrow mem or a broad nun,87 • Position 8. A yod (photos 579, 584, 641, 671, 684–87,
and the leftmost stroke interpreted as either a word 691, 692, 693, 694, 2907), followed by a word divider
divider—in that case, followed by a waw in position (photo 684).
5, or as part of the qop in that position. (3) The shaft • Position 9. A he (photos 009, 030, 671, 2907).
and all three strokes could constitute a mem, in that • Position 10. An alep (photos 009, 671, 2907). Of the
case followed by a waw. expected word divider, a remnant may be visible ei-
ther between the tips of the alep and lamed, or a little
For position 4, on formal considerations, mem or nun farther down (same photos); in the transliteration, we
are equally possible; for position 5, a waw is our first have taken the liberty to mark it as certain.
choice, a clumsy qop (compared with position 4 in line 9) • Position 11. A lamed (photos 030, 671, 2901).
the second.88 We opt for a waw, as it allows the preceding • Position 12. A kap (photos 633, 636). In most other
stroke to be identified as a word divider, which affords in photos (e.g., 2901, 2902), the letter looks like an over-
turn a plausible reading of part of line 6. sized pe—the middle “finger” is mostly broken away.90
• Position 13. A bet (photos 030, 671, 2901). Intact ex-
• Position 6. A ḥet(?). If so, it is placed unusually low on cept for slight breakage at top right.
the line. Lidzbarski noted that the present remains are • Position 14. A bet and a word divider (photos 030, 671).
incomplete and practically unidentifiable, hesitatingly • The next six letters read clearly ʾrmly.
proposing a ṭet,89 which we, in turn, do not see—there • Position 15. An alep (photos 030, 633, 634, 663, 671–
are no curved lines in the letter. A ḥet seems to us 73, 2907).
possible, assuming with Lidzbarski that the letter is • Position 16. The resh seems quite certain (photos 030,
incomplete: it now ends in an oblique break that elim- 576, 633–34, 663, 671–73, 2907), yet this is the least
inated its bottom half, as best seen in photos Istan. 346 clear letter among positions 8–19, and an alternative
and 2041, and our 029, 635, and 641. A small dot on gimel or waw (not drawn) is not to be wholly excluded.
the right, a little lower than mid-line, was regarded by • Position 17. A mem (photos 559, 673, 2901). “. . . vor
Lidzbarski as either a word divider or a remnant of ‫ לי‬kein Worttrenner” (ESE 3: 199); we agree.
the broken letter itself (see n. 89), yet in his transliter- • Position 18. A lamed (photos 559, 2901).
ation he marked it as a divider, while we—assuming • Position 19. A yod (photos 559, 578, 2901).
the preceding letter to be the copulative conjunction
waw—follow his latter choice of a letter remnant. Line 7
• Position 7. A nun, Lidzbarski’s mem (transliteration Lidzbarski counted seven letters to the right of the bet
only; no justification in the text). See Lidz. pl. 14 and preceding rkbʾl, we eight: he regarded our positions 6 and
photos 584, 641, 668, 671, 672, 673, 684–86, 690, 692– 7 as one letter, for which see below and n. 92. And in case
694, and 2907. position 1 turns out to be a phantom, we will count seven
letters too, but differently.
Positions 8–19 and their word dividers are generally
clear (with the possible exception of 16): y•hʾ•lkbb•ʾrmly. • Position 1. A samek(?). Only three parallel horizontal
Our identifications follow Lidzbarski’s. strokes at the right edge are clear in most photos. (In
87 The shaft with its narrower tip and two strokes to its left, packed
a he, one would expect these strokes to be tilted bot-
closely together, might pass for an awkward nun in addition to the more
tom left.) The vertical of the letter may be apparent
obvious mem. Lidzbarski favored a mem: “Der . . . Buchstabe ist ziem- in photos 637, 638, 639, 640 and 2906, but it is feeble
lich sicher ein Mem, aber unbestimmt ist es, wie weit der Kopf nach and far from certain. Alternatively, this is a phantom,
links reicht: ob der letzte Strich vor dem schrägen Schaft noch zu ihm the three horizontals part of the texture of the basalt.
oder zum zweiten Zeichen gehört. Ein Worttrenner kann der fragliche • Positions 2–3. Among several possibilities we thought
Strich nicht sein, da er zu lang ist” (ESE 3: 198–99). Yet the length of
that stroke does not preclude its identification as a word divider, and
lamed-alep the likeliest, reassigning Lidzbarski’s word
this option, with its ensuing reading, is to us preferable to attaching the divider to the lower stroke of the alep’s chevron (e.g.,
stroke to the following letter. Lidz. pl. 14 and photos 637, 638, 691, the latter repro-
88 Lidzbarski wrote on position 5, “Beim . . . Buchstaben muß man
ducing the alep’s shaft best).91
zwischen ‫ כ‬und ‫ ו‬schwanken. . . . nicht etwa ein ‫ ק‬. . . , wofür der Schaft
zu stark nach links ginge” (ESE 3: 198–9). We share neither Lidzbarski’s
potential acceptance of a kap nor his reason for rejecting a qop. 90 Inphoto 671, the head of the letter even appears closed as in a
89 “Das folgende nach unten spitz zulaufende Zeichen kann nicht resh, certainly a false impression.
vollständig sein; unten ist ein Stück abgeschliffen. Vielleicht hat ur- 91 Another alternative may be zayin-alep, the small zayin formed
sprünglich ein ‫ ט‬dagestanden. Der Punkt rechts kann ein Worttrenner by combining Lidzbarski’s lamed and word divider differently (same
[between our positions 5 and 6] oder auch ein Rest des fehlenden photos). But this is even less likely, for it would place part of the alleged
Stückes sein” (ESE 3: 199). zayin inside the alep’s head.


• Positions 4–5. Uncertain: nun-waw or mem-nun (Le- • Position 6. A yod? (photos 011, 579). In other photos
maire), mem-pe or mem-divider-pe or nun-divider-pe (e.g., 636, 641, 663, 671), all that can be seen of the
(the latter probably preferable—Sass). Photos 553, same letter is a nearly vertical line.
576, 579, 691, and 696 show why it cannot be decided • Position 7. Perhaps a waw (photos 636, 641, 663, 671)
whether the middle stroke of this cluster belongs to or a nun (011, 579), but there is no certainty. Photos
position 4 or 5 or constitutes a word divider. Com- 554, 555, and 636 give the impression of a sole letter
bined with the letter in position 5, the said stroke occupying positions 6 and 7, a vast ḥet—indeed, Lidz
renders a Y-shape that can be interpreted as waw. Yet barski’s first assumption, subsequently discarded—see
the ensuing shape may be too large compared with n. 92. As Lidzbarski remarked, the letter would be
most other letters in line 7, hence the alternative sug- broader than usual; in most other ḥets, their height
gestion of a pe, with the stroke at top right belonging exceeds their width.
either to a divider or to position 4 (i.e., either a broad • Position 8. Unidentified. Of Lidzbarski’s alep we see
mem, or a narrow mem and a divider, or a nun and a no trace. In photos 032, 579, 585, and 689, and several
divider). We understand besides what led Lidzbarski others, one might make out a rounded shape with a
to propose a shin or ṣade for position 5, but both seem
improbable to us.92 remarked: “Die Gruppe, die ich durch •‫א‬-‫א•( ש‬-‫ )צ‬wiedergebe, ist ganz
verworren. Ich habe anfangs ‫ שחי‬hingeschrieben, aber die für ein ‫ח‬
92 Each was doubtful to Lidzbarski too, and (referring to our posi- beanspruchten Züge sind dafür zu breit, stehen auch zu schräg” (ESE
tions 5–8, which he regarded as three letters followed by a divider) he 3: 199).

‫ש א‬ • ‫ר כ ב א ל‬ ‫• ב‬ ‫א‬ ]‫[ח‬/– ‫ש‬/‫צ‬ ‫נ‬ ‫א‬ • ‫ל‬ ‫ס‬

‫ש א‬ • ‫ר כ ב א ל‬ ‫ב‬ – ‫נ‬/‫ו‬ ‫פ י‬/‫נ‬/‫נ • ו‬/‫מ‬ ‫א‬ ‫ל‬ ‫ס‬
16 15 ? 14 13 12 11 10 9 8 7 6 5 ? 4 3 2 1
cross inside, a candidate for ṭet (see n. 106), and in • Position 11. A kap (Lidz. pl. 15 and photos 008, 010,
photos 030 and 636 a yod? (Or he? Not drawn.) These 033, 635, 663, 671, 694, 2901, 2904, 2905, 2906).
are the same remains, but with a different empha- • Position 12. A rather broken bet, best in Lidz. pl. 15
sis, and probably illusory (yod was also considered and photos 009, 030, 031, 636, 672, and 694.95
by Lidzbarski and then rejected—see n. 92). In his • Position 13. A much damaged alep (photos 556, 2900,
transliteration, Lidzbarski indicated a word divider 2904, 2906).
between positions 8 and 9, as had Bossert, certainly • Position 14. A lamed (photos 2901 [illustrating letters
an expected feature, but we did not see it (photos 030, 8–15], 2907). There is room for a word divider in the
671, 693).93 space to the left: Lidzbarski indicated one in his trans-
literation and Bossert drew one. Should it exist, it is
Indeed a plethora of possibilities. nearly obliterated (Lidz. pl. 14 and photos 031, 632,
Positions 9–16 are clear (Lidz. pl. 15 and photo 2901), 633, 2907, etc.), but remnants are possibly visible in
the text intelligible:94 brkbʾl•šʾ. . . . Our identifications Lidz. pl. 15 and photo 2901.
follow Lidzbarski’s and Bossert’s. • Position 15. A shin (photos 631, 632, 2907). The
middle tip of the shin is topped by a blob that seemed
• Position 9. A bet (photos 554, 575, 636, 684, 693). to us not to belong to the letter; we have left it out of
• Position 10. A resh (Lidz. pl. 14 and photos 684, 693). the drawing.
• Position 16. An alep (photo 2907).
93 One of the traces that we assume belongs to the largely oblit-
erated letter in position 8 could easily be imagined to constitute the
remains of a word divider instead. 95 Inphotos 010 and 637, the letter may look like a circle with a
94 Only the word divider is uncertain. central dot, certainly deceptive (see n. 106).

“Hinter ‫ אלה‬Z. 5 und ‫ שא‬Z. 7 sind keine Punkte zu • Position 12. The alep is damaged; clearest in photos
sehen, danach sind beide Wörter mit dem folgenden ‫ין‬ 030, 636, 673.
zu verbinden” (ESE 3: 200; see also p. 203 with n. 2). For • Position 13. A yod. To us the letter seems effaced, yet
šʾ this is certain, for ʾlh possible. its presence here in šʾyn is beyond doubt, and rem-
nants may be discernible in photos 009 and 641. An
Line 8 approximation of the letter was drawn by Bossert (our
Bossert (1932) improved Lidzbarski’s identification of Fig. 9), probably on account of its recurrence in the
positions 10 and 13/14. On the other hand, he omitted formula l-DN šʾyn lym rather than of the actual re-
the letter in position 6, on the identification of which we, mains on the stone at this point.98
in turn, differ from Lidzbarski. There is general agree- • Position 14. A nun. A blow to the stone has eliminated
ment on the identification of the remaining 14 letters and part of the nun’s leg, unless traces can be seen just to
5 word dividers in line 8—which are, indeed, the clearest the left of this spot (photos 009, 635, 636, 641). The
in the inscription. following word divider is clear in photos 009 and 673,
among others.
• Positions 1–2. The yod is squeezed in at the beginning • Position 15. A lamed (Lidz. pl. 14 and photos 009, 031,
of the line, but quite legibly. The nun is damaged, but 554, 634, 673).
its restoration is certain on the strength of the par- • Position 16. A yod (photos 031, 634, 673).
allel in positions 13–14 (Lidz. pl. 14, 15 and photos • Position 17. A mem, probably followed by a word di-
637–640). The word divider is clear (Lidz. pl. 14 and vider (photos 558, 584, 663).
photo 638).
• Position 3. A lamed (Lidz. pl. 15 and photos 008, 010, Line 9
556, 558, 637, 638, 639, 640). • Position 1. We opt for a waw against Lidzbarski’s kap
• Position 4. A yod (Lidz. pl. 15 and photos 008, 010, or shin preference99 and the taw or ṣade that we con-
556, 558, 637, 638, 639, 640). sidered initially.100 The lighting of Lidz. pl. 13 and
• Position 5. The mem is quite worn, but still recogniz- photos 011, 028, 029, 585, and 689 results in a more or
able (Lidz. pl. 15 and photos 008, 010, 556, 558, 637, less Y-shaped letter missing much of its leg.
638, 639, 640). “Vor ‫[ זב‬Lidzbarski took position 6 • Position 2. Probably a bet (photos Istan. no num-
to be a zayin] könnte ein Punkt stehen” (ESE 3: 199). ber and our 640, 2901) and, if so, it differs, by being
Indeed, a word divider is to be faintly seen in Lidz. pl. slightly open, from the other clear bets in the inscrip-
15 and photos 010 and 638. tion, 17 in all. In photos 011, 557, 579, 585, 689, 696,
• Position 6. A Y-shaped waw (photos 030, 031) or and 697, the letter may look more like a pe. Lidzbarski
Ч-shaped (558).96 has listed both possibilities in his transliteration but
• Position 7. A bet (photos 579, 635). refrained from commenting on them in the text. We
• Position 8. A kap (photos 010, 577, 666, 669). The dot opt for the bet, forming part of what seems to us a
above must be accidental. plausible reading.
• Position 9. A damaged bet (photos 635, 637). • Position 3. Probably an oblique-headed mem, much
• Position 10. This bet and the following word divider of its three zigzags worn down (photos 008, 603, 637,
are worn down and almost fused together. They are
best visible in Lidz. pl. 14. The wear causes the ensem- 98 The old Istanbul Museum photographs (no number, and
ble to look quite like a he in certain photos (e.g., 636).97 2042/2072) show that the stone was as effaced at this point in Lidzbar-
• Position 11. A shin (photos 030, 636, 673). ski and Bossert’s time as it is today.
99 Both of which he himself doubted: “Das erste Zeichen ist nur
halb erhalten, und es ist zweifelhaft ob es ‫ ש‬ist. Außerdem könnte es

96 At one point, we considered the option of a vertical stroke be- nur ein ‫ כ‬sein, aber auch dafür wäre die Form ungewöhnlich” (ESE 3:
tween two dividers (photo 029): the numeral 1? But we later discarded 199). To us, the single v-shape visible is too close to the edge of the stone
it. As noted, Lidzbarski was categorically for a zayin (“Das ‫ ז‬ist sicher”; to permit a second one to the right, now obliterated, for a shin (on the
ESE 3: 199); Bossert left an empty space. left the next letter is too close); besides, the letter has a leg, now mostly
97 This he-like appearance explains why Lidzbarski considered obliterated. “‫ נב‬steht nicht da” was another remark, somewhat vague, of
identifying position 10 as a ḥet, itself not too far removed from he in Lidzbarski’s (ESE 3: 199) on positions 1 and 2, all the more ambiguous
shape (bet was his second choice): “Recht zweifelhaft ist das ‫ ח‬hinter ‫כב‬ as one of his options for position 2 was, after all, bet.
. . . Es könnte an sich ein ‫ ב‬mit einem Punkte dahinter sein, vgl. ‫ כבב‬in 100 In most photos (e.g., photos Lidz. pl. 15, Istan. no number, and
Z. 6, aber zwei so verschiedene ‫ ב‬können nicht neben einander stehen” our 557, 689, 696, 697), an asymmetrical v-shape with its right arm
(ESE 3: 199). Evidently they can, but Lidzbarski has not recognized longer is all that remains visible of the letter, and this may alternatively
here the divine name Kubaba; this was left to Bossert in 1932. See also be restored as a taw. We prefer the waw, which may yield a meaning-
Lidzbarski’s hesitations on positions 3–4 in line 5. ful reading.



• ‫• ל י מ‬ ‫א ש‬ ‫ב • ש‬/‫ח‬ ‫כ ב‬ ‫י מ • ז ב‬ ‫ל‬ • ‫ י נ‬Lidzbarski

• ‫א י נ • ל י מ‬ ‫• ש‬ ‫ב‬ ‫כ ב‬ ‫ב‬ • ‫י מ‬ ‫ל‬ • ‫י נ‬ Bossert
• ‫א י נ • ל י מ‬ ‫• ש‬ ‫ב‬ ‫כ ב‬ ‫י מ • ו ב‬ ‫ל‬ • ‫י נ‬
17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
689). The leg may be likewise oblique and slightly scheinlich” (ESE 3: 199).102 We see no divider between
curved (photo 689). Yet Lidz. pl. 14 and photos 555, positions 6 and 7.103
556, and 557, and most others, show why Lidzbarski • Position 7. A lamed (photos Istan. 346 and our 029,
identified a nun here: indeed, the head of the letter 558, 584, 585, 663, 672, 673, 690, 2920). The rounded
looks like a nun’s in them,101 if as broad as a mem’s, shape apparent in several photos (e.g., 667, 690) is
and the leg looks vertical (in photo 008, both optional probably the result of lighting, yet the letter may in-
legs are seen together). While a nun cannot be defi- deed be less acutely angled than most other lameds in
nitely excluded, we tentatively favor the mem in view our inscription.
of the ensuing mqm, which fits our proposed reading, • Position 8. The kap is partly broken away, and no
while nqm does not. photo shows it in its entirety. Its identification from
• Position 4. A qop (Lidz. pl. 14 and photos 557, 603, the stone and from several photos combined is rea-
635, 637, 638, 639, 640, 689, 2900, 2901, 2905, 2906). sonably certain, however (photos Istan. 346 and our
• Position 5. This mem is much damaged; its remains 004, 009, 029, 030, 584, 632–41, 671–73, 690, 2907).
can best be seen in photos 556, 635, 637–39, and 669. • Position 9. Quite recognizable as a yod, yet the letter
There follows a divider, located closer to the next is so damaged (Lidz. pls. 14, 15 and photos 009, 030,
letter. 558–59, 577, 584, 632, 633, 634, 635, 636, 638, 640,
• Position 6. The mem is damaged, yet not in doubt— 2901, 2907, 2920) that it is hard to tell whether the sole
see photos Lidz. pl. 14, Istan. 346, and our 556, 558,
559, 576–78, 584, 603, 635–41, 666, 667, 669, 674,
102 Lidzbarski continued: “‫ מלכי שמאל‬ist gänzlich ausgeschlossen.”
690, 695, 2900, 2901, 2905–7, and 2920. Lidzbarski
To us, mlky•š. . . is fairly clear (see main text), but we agree that the
hesitated between mem and nun and added: “Vor ‫לכי‬ three following letters do not read mʾl.
ist ein Punkt sichtbar, daher die Lesung ‫ מלכי‬unwahr 103 What looked to Lidzbarski like a word divider (e.g., Lidz. pl.
14 and photo Istan. 346) is probably the tip of the mem’s own head,
or just the right-hand edge of the depression enclosed by the lamed. It
should be noted that several of the mems in our inscription are so worn
that a nun followed by a word divider becomes an almost automatic
101 What might have strengthened the impression of a nun, the alternative (see, for instance, position 13) but not, apparently, in the
down-pointing tip at the bottom left end of the head, probably does present case; we regard the sequence wbmqm•mlky as reasonably well
not belong to the letter; without it, the head becomes a mem’s. established.




‫• י‬/– ‫מ‬ ‫י‬/‫א‬ • ‫צ‬/‫נ • ל כ י • ש א‬/‫מ‬ ‫מ‬ ‫ק‬ ‫נ‬ ‫פ‬/‫ש ב‬/‫כ‬
‫י‬ ‫נ • ל‬/‫מ‬/‫כ‬ • ‫י‬ • ‫ת‬/‫ל כ י • ש א‬ ‫מ‬ • ‫מ‬ ‫נ ק‬/‫מ‬ ‫פ‬/‫ב‬ ‫ו‬
15 14 ? 13 ? 12 ? 11 10 9 8 7 6 5 4 3 2 1
clearly visible left horizontal is the top or the middle tioned, as in photos 008, 558, 559, 577, 584, 663–65,
one—hence, the two alternative restored drawings. A 690, and 2900. The shaft appears to lean top right in
word divider follows (photo 009). these photos—as befits a taw—but this in turn could
• Position 10. The shin is clear in many of the photos be due to the texture of the stone (other photos show
(e.g., 558, 635, 671–73, 690). The space between po- a vertical shaft, as noted). Lidzbarski’s transliteration
sitions 10 and 11, larger than usual, may be due to has alep or ṣade followed by a word divider; he had
a preexisting depression in the stone that the mason not discussed this letter in his text.104
avoided.
Letters 12–15 are rather squat and close together, pos-
sibly due to the mason’s reluctance to split a word be-
Positions 11–13. Some of the restorations of the next
tween the non-contiguous lines 9 and 10.
three letters have an optional word divider after each let-
ter, though only one of them can obviously be real. We
• Position 12. A yod (photos 030, 634, 673), whose
opt for the last one, but this is not absolutely certain.
plump appearance in several photos (e.g., 556, 558,
690) could be due to damage. If our reading is ac-
• Position 11. Either an alep alone, or an alep and a word
cepted, what may otherwise have been interpreted as
divider, or a taw and a word divider are possibly re-
a word divider is part of the yod’s middle horizontal
storable: The restoration of an alep alone (after photos
(photos 556, 558 again). On Lidzbarski’s choice, see
632, 635) results in the reading šʾyn ly. The top of this
n. 105.
alep’s vertical would be uncommonly bent to the left,
• Position 13. A nun and a word divider? These are pro-
though in our inscription, this letter has a number of
posed in light of the ensuing reading (see position 11).
variants, with letter 7 in line 10 somewhat similar. The
remains on the stone may likewise permit restoring a
strictly vertical alep, with a word divider to its left, the
latter the same as the top of the alep’s vertical in the 104 The shaft, appearing nearly vertical in certain photos (Lidz. pl.
foregoing restoration (photos 009, 584, 632, 634, 635, 15, Istan. no number, and our 557, 632, 634, 641, 671–73, 696, 2901,
673; this corresponds to Lidzbarski’s identification— 2905, 2907, 2920), would have led us to prefer a ṣade, but the horizontal
wavy line of that letter is not there. Indeed, the eye may merely con-
see below). The divider would eliminate the said nect in different ways a mix of wear and letters’ remains here, and thus
appealing reading, however. Another alternative is photos 028, 029, and 033 give the unlikely impression of a large yod,
a worn taw followed by the word divider just men- its top horizontal overhanging the divider mentioned in the main text.

A worn mem (Lidzbarski),105 or a kap, in case what • Position 4. A he (photos 545, 547, 571).
we have taken for a divider is in fact part of the letter, • Position 5. A nun (photos 013, 014, 545–47, 566, 567,
could fit the same remains (cf. line 4, position 16). The 571, 572, 587, 604, 607, 609, 659, 2913, 2915).107
rounded, squat shape is apparently clear (best visible • Position 6. A yod and word divider (photos 014, 545–
in photos 008, 556, 558, 584, 665, 690); it differs from 47, 571, 572, 587, 607, 659, 2914, 2915). The letter is
the Zincirli norm for nun, mem, and kap alike, but more compact than usual, but unmistakable.
this could be due to the compact execution of letters • Positions 7, 8, 9. The next three letters and following
12–15 as a whole.106 word divider are clear in photos 014, 587, 607, 659,
• Position 14. Possibly a lamed. Photos Istan. no num- 2912, and 2915. In his transliteration, Lidzbarski left
ber and our 030, 558, 584, 634, 2907, and 2920 might out the word divider following position 9, regarding
display the remains of a small lamed, worn almost flat. it as part of position 10 in order to make that letter
Lidzbarski did not see it: “Zwischen dem folgenden ‫מ‬ into a kap (see below on position 10); but in the text,
[position 13] und ‫[ י‬position 15] steht vielleicht nur he mentioned the alternative of a divider,108 which we
ein Punkt” (ESE 3: 199). We, too, discovered it only at certainly prefer. We have no further comment on the
quite a late stage, when we had acquainted ourselves alep and shin in positions 7 and 8.
well with the writing. • Position 9. A mem? The letter was identified as mem
• Position 15. A yod (same photos). Damaged but quite by Lidzbarski, followed by us on account of the ensu-
clear, similar to the letter in position 12. ing reading ʾšm, “the name of.”109 Yet such a shape,
with its vertical tapering leg and unique head (clear in
Line 10 most photos, e.g., 013, 571, 614, 627, 659, 2912, 2915)
Although it is the continuation of line 9, line 10 be- is quite atypical (see also n. 51). Indeed, Lidzbarski
gins around the corner from line 7 (Fig. 1c) if the waw is had to “improve” the shape in his drawing (Fig. 8)
considered its first letter. in order to render it closer to the norm. Or could the
Pre-position 1. Line 10 was thought by Lidzbarski to wedge-shaped leg and middle “finger” of the letter be
begin with the waw of wzyhny, and he was probably right. the result of damage, or an optical illusion? In photos
But in several photos (566, 659, 2912, 2915), it looks as if 017, 548, 550, 566, 567, 589, 609, 617, 629, and 662,
this waw is preceded by a lamed and a large word divider. the letter may be restored as a damaged resh, though
Although much more visible than, say, the remains of with less credibility than its identification as a mem.
the lamed we propose in line 9, position 14, it is probably A resh would, moreover, furnish a plausible alterna-
illusory; certainly so if it is accepted that the end of line tive reading, ʾšr, “the [resting] place of.” As noted, a
9 is the end of a word. Once again, this phenomenon word divider follows.
reminds us of how precarious much of the reading of the
Ördekburnu inscription is. Positions 10–13 are much damaged, as can be seen
Positions 1–6. The majority of the letters of wzyhny, clearly in photo 2912 (illustrating letter 11). A personal
followed by a word divider, are clear in most photos. The name may be expected to follow if ʾšm or ʾšr is read in
best views of the individual letters are as follows: positions 7–9. Lidzbarski (ESE 3: 200) reached the un-
reasonable conclusion that the sequence in positions
• Position 1. A waw (photos 014, 566, 567, 609, 659, 10–12, read kyn by him, was the alphabetic rendering
2914, 2915). of the name of Kulamuwa’s father, Ḫaianu in Assyrian.
• Position 2. A zayin (photos 545, 547, 604, 2915). While it is not impossible to understand the three let-
• Position 3. A yod (photos 545–47, 571, 587, 2914, ters as kap, yod, and nun, the identification of Ḫaianu’s
2915). name in the string kyn is out of the question in view of
the alphabetic writing ḥy or ḥyʾ in Kulamuwa’s inscrip-
105 Thus, in the transliteration; also in the text, ESE 3: 199, though tions (on the latter, see Tropper 1993: 30, 32, 38, 51). All
without an explanation: “Vor dem ‫ מ‬am Ende schrieb ich anfangs ein the same, this understanding was often repeated, e.g., by
‫י‬. . . .” See also n. 103.
106 A circle that could have been identified as an ʿayin is to be seen
instead in photos 554, 575, 584, 664, and 665, but it is illusory—com-
prising, in fact, our nun and divider (or mem/kap alone) and the right-
hand arm of the following (suggested) lamed. Other ostensibly rounded 107 The hole to the right of the letter is probably natural. In any case,
shapes in our inscription that may be taken for ʿayins (3/12, 4/6, 4/11, as shown by photos such as 013, it has not destroyed a purported top-
5/9, 7/7, 7/11) and seem, like the present one, to be located around a right part (which could have suggested a kap).
flat, circular, or oval depression are all misleading. Two or three of these 108 “Zweifelhaft, ob zwischen ‫ אשמ‬und ‫[ כין‬latter our positions
circles may even give the impression of enclosing a stroke or a cross 10–12] ein Punkt steht” (ESE 3: 199).
shape (4/6, 7/8, 10/11), phantom ṭets. 109 Thus already Landsberger 1948, quoted in n. 27 herein.






‫ל ה‬ ‫א‬ ‫ש‬ ‫מ‬ ‫ש‬ • ‫נ ב‬/‫כ‬ • ‫ת ה א‬ ‫מ‬ ‫נ‬ ‫י‬ ‫• כ‬ ‫מ‬ ‫ש‬ ‫• א‬ ‫נ י‬ ‫ו ז י ה‬
• ‫ל י‬ ‫ע‬/‫י‬ ‫ל ו‬/‫ח י‬/‫ה‬ • ‫ש‬ ‫ב‬ ‫נ‬ • ‫ס ת ה י‬/‫ע א‬/‫נ‬/‫י ל‬/‫ט‬/‫פ ח‬/‫כ‬/‫ר • ה‬/‫מ‬ ‫ש‬ ‫• א‬ ‫נ י‬ ‫ו ז י ה‬
25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
Tropper (1993: 6),110 and also influenced the dating of traces and eliminating others would result in a shape
the mortuary scene (see “The Stele in Sam’alian Geog- resembling that of a yod, which might in turn yield
raphy and History” below). On our alternative, very un- the conceivable, if uncertain py-, Piya-X, mentioned
certain py. . . , “Piya-X,” see “Tentative Translation and above. In different lighting (photos 548, 570, 592, 610,
Commentary.” 660, 661, 662, 663), the same traces may look mid-
way between a square and a circle, with a cross shape
• Position 10. Possibly a pe or he, less likely a kap. In inside—a candidate for a ṭet. All these are highly un-
a certain lighting, the letter is reminiscent of a pe, certain: what we see are just meager remnants, while
its leg curved (photos 017, 548, 549, 563, 592, 610, most of the letter no longer exists. Lidzbarski too
658–62, 2911, 2912, 2914). In different lighting, it proposed a yod, but his letter was put together out of
may resemble an acutely angled he (photos 013, 014, entirely different remains, barely distinguishable in
545, 571–73, 587, 612, 613, 614, 615, 623, 625–27).111 photos 014, 546, 563, 571, 573, 626, and 2914, which
Viewed another way, one may see the thin lines of a we cannot accept.113
nearly effaced kap in the same photos. The different, • Position 12. A lamed, nun, or ʿayin. Lidzbarski was for
bold kap sketched by Lidzbarski definitely does not a nun,114 and indeed, if one assumes that its leg has
exist (see above on positions 7, 8, 9).112 flaked off (remnants possibly visible in photos 574,
• Position 11. Damaged beyond recognition. In photos 614, and 625), the letter could originally have resem-
2911 and 2912, a squarish shape could, with some ef- bled the one in position 5. Alternatively, the presumed
fort, pass as a worn ḥet. Admitting only some of these nun’s head may constitute the entire letter, and if so—
which is by no means guaranteed—it becomes a can-
110 “Urheber der Inschrift ist wahrscheinlich König Ḥayyā von
YʾDY/Samʾal. Die Inschrift wäre dann in das zweite Viertel des 9. Jh. 113 Lidzbarski has prettified the form of that yod in his drawing
zu datieren. . . .” (Fig. 8). If anything, this shape would resemble somewhat the later
111 The obtusely angled he ostensibly visible in photos 550, 566–68, Phoenician yod, rounded and turned 90º counterclockwise, which is
589, 593, 604, 609, 628, and 629 appears to us illusory. obviously out of the question here.
112 A nun, Lidzbarski’s second choice, is, in its turn, not too far 114 Yet Lidzbarski has drawn this curvilinear letter as if it were
removed in shape from our proposed pe. straight-lined and angular (Fig. 8).

didate for a small, rounded lamed (slightly open circle, between nun and kap, and in ESE 3: 199, a propos let-
same photos as for nun; somewhat like position 14 in ters 17 and 18, he wrote: “Hier ist ‫ כב‬ebenso möglich
line 5),115 or an ʿayin (practically closed circle, e.g., wie ‫נב‬.” In the transliteration, he wrote a ‫ נ‬with a dot
photos 592, 607, 622, 657, 658, 659). above.
• Position 13. Uncertain. With hesitation, we propose an • Positions 18 and 19. The bet and shin are particularly
alep (Lemaire) or a samek (Sass). In photos 627 and clear in photos 016, 550, 569, 604, 606, 611–15, 618,
661, a sideways v-shape, the head of a possible alep, 622–24, and 630. The shin seems to be followed not
is visible. Other photos (550, 568, 570, 592, 609, 610, only by a dot but also by a slightly curved line farther
617, 2915) seem to suggest a worn-down head of a left. We could not interpret this line as part of the next
samek,116 and also, with a little more good will, a pos- letter, nor explain it otherwise. Accidental?
sible leg, but all this is far from certain. No divider is
visible to the left; indeed, there is hardly room for one. Positions 19–25, seven letters in our count, were read
• Position 14. A taw, inscribed within a depression. This šmš ʾlh (six letters), “the god Shamash,” by Lidzbarski, a
was Lidzbarski’s reading, and the letter is clearly vis- beautiful reading that cannot be maintained.120 The ini-
ible in most photos (e.g., 658, 662, 2911).117 tial shin and penultimate lamed are there, but the word
• Position 15. The he is undeniable in most photos (e.g., division, as well as the reading of the other letters, even
551, 566, 571, 606–8, 661, 662, 2914) but badly pre- their count, are far from clear: Lidzbarski saw two letters
served, an impression given also by Lidzbarski in his between the shin in position 19 and his alep in our posi-
drawing (Fig. 8). The leg is bent in the same manner tion 23 (not counting the two said letters themselves),
as in the last letter of line 5. while we think there are three, plus a possible divider.
• Position 16. A yod? Followed by a word divider? Or The shapes that Lidzbarski deciphered as mem-shin of his
else the two may have constituted a single letter, now šmš thus seem in the photograph to be too large if mea-
worn away beyond recognition (photos 566, 606–8, sured up to the rest of the line (see n. 122). This disagree-
621, 658, 661, 662, 2914).118 ment in size must have troubled Lidzbarski too: When
one compares his drawing (Fig. 8) to the photographs,
Positions 17–19. In our mortuary context, reading the it appears that he has attempted to tone down the dis-
three letters nbš is obviously welcome, and the word di- crepancy by sketching the preceding letters thicker than
vider identified by Lidzbarski between the bet and shin is they really are.
doubtful at best.
• Position 20. Unidentified. A he, gimel, nun, or pe (lat-
• Position 17. The nun is not very clear (photos 013, 547, ter three not drawn) (Lemaire), or a ḥet (Sass) (photos
564, 571, 573, 574, 607, 616, 630), but seems the likeli- 013, 016, 017, 547, 550, 571, 662, 2911, 2914).121 One
est choice.119 Lidzbarski’s drawing (Fig. 8) is a cross oblique stroke on a shaft to its right is clearly visible in
many of the photos—hence gimel, nun, or pe; and in
photo 550, we believe we see all three obliques—hence
115 Ifa lamed, the letter would be quite different from the clear
lamed in position 24.
he or ḥet. In any case, Lidzbarski’s drawing of a head as
116 The head of this proposed samek is indeed reflected in the three that of a mem (Fig. 8) cannot be reconciled with the
vertical bow-shapes at the top left of Lidzbarski’s unlikely drawing of remains on the stone.
a mem with a head like no other, and oblique leg, Fig. 8. On the stele, • Position 21. Much obliterated. A small yod may be
we could not see this mem. Similar remains are perhaps to be seen in discerned in photos 548, 549, 561, 562, 619, 620, 662,
line 4, position 7.
117 What may oppose this, but not gravely, is the vertical stance: and 2914 or, viewed differently, a small angular lamed;
True, most contemporary Aramaic and Phoenician taws are inclined but each is uncertain. This is the right-hand part of
top right, but vertical ones are documented occasionally. The only al- Lidzbarski’s shin.
ternative that comes to mind, a waw, is not without its own difficulties: • Position 22. A waw may be suggested by the light-
the clear right-hand half of the taw’s horizontal will have to be ignored, ing of most photos (e.g., 016, 017, 549, 551, 560, 593,
and a small vertical, now absent, restored at left.
118 Lidzbarski identified the letter as an alep: The presumed “chev- 617–20, 662, 2911, 2912, 2914), but in others (e.g.,
ron” of the letter is there (we regard it as the would-be bottom-right
end of the yod), but the top and bottom verticals drawn by Lidzbarski 120 Could it be that Lidzbarski first identified two or three letters
(Fig. 8) are not. The alleged bottom vertical is, in fact, sunk rather than and then (probably inspired by rkbʾl ʾlh in line 5) he had the “šmš ʾlh”
in relief, a damage to the stone (photos 658, 661, 662), firmly ruling idea, adapting to it the rest of the letters?
out the alep option. 121 In other photos, the lighting causes the letter to look like an
119 In other photos (e.g., 661, 662), the head of the letter may look obtuse-angled or right-angled he (562, 566, 567, 604, 609), but this
closed, but since this does not render any recognizable letter shape (resh seems illusory to us. A somewhat similar phenomenon can be observed
would have had a vertical, unbent leg), we think it must be illusory. in position 10.

012, 013, 565, 616) the letter looks different. This is

the left-hand part of Lidzbarski’s shin.122
• Position 23. Illegible. Several photos (012, 561, 562,
565, 616) show a shape resembling a tiny rhomboid
ʿayin. The surface above is probably intact, but just
below there is a sunk spot due to damage (e.g., photo
2914), so that the presumed ʿayin may be only part
of the original letter, now mostly gone. Indeed, one
of us (Lemaire) distinguishes a possible small yod in
photos 562 and 616 by admitting three of the rhomb’s
four sides and some additional traces below and to the
right. It is impossible to know whether one of these
readings is right or whether both are wrong. Lidzbar-
ski was for an alep and, looking at photos 016, 017,
549, 551, 619, and 620, one can be made out with
much good will, but it may be illusory. The shape that
Lidzbarski has drawn (Fig. 8) is far too optimistic,
however.
• Position 24. A clear, rounded, full-sized lamed (e.g.,
photos 562, 2911).
• Position 25. Remains of an obliterated letter, perhaps
the bottom half of a yod (e.g., photos 017, 618, 2911,
2912).123 A word divider probably follows.
• Post-position 25. Following a space, a shape resem-
bling a thin-lined, round-headed yod is visible in
photos 016, 017, 593, 2911, 2912, and 914, its left side
merging into the unworked surface of the stele. Or
else the same may be perceived as just a circle, an ʿayin
(not drawn). All this is probably accidental, however.
A continuous transliteration. As presented here (in belong to the Sam’alian personal pronoun of the first
col. 2, this page), gray signifies more conjectural identi- person singular ʾnk, already evoked by Lidzbarski (ESE
fications. For letter 12 in line 3, letters 2–3 in line 6, and 3: 197–98),124 expected here on comparison with the in-
letters 11 and 13 in line 9, the transliteration below can- scriptions of Hadad and Katumuwa. If this was so, the
not reflect the problematics in full, and the line-by-line personal pronoun will have been followed by a personal
discussion has to be consulted. name, possibly also by br/brt and the patronymic,125 and/
or a title,126 or, in case the deceased was a woman, by
Tentative Commentary and Translation “wife of PN.” Yet the state of the text prevents us from
making a concrete suggestion founded on the traces of
Given the find-spot of the stele, the comparisons in the letters.
this commentary are first and foremost with the other
inscriptions found in the land of Sam’al. Line 2
Line 1 The existence of a letter in position 1 is doubtful,

and the following two or three letters are of uncertain
The line is so damaged that not even a single word reading.127 Between two word dividers, the three let-
can be recognized. The possible nun in position 2 could
124 In case the inscription begins with position 2 (see discussion of
122 Photos 015, 566, 567, and 2913 show what led Lidzbarski to position 1 in “Identification . . .”), nṣb becomes an option, or else mqm
read a large shin occupying our positions 21–22. In a similar manner, if the reading of mem is preferred.
Lidzbarski read together as a large shin letters 13 and 14 in line 8—in 125 As in Hadad: ʾnk pnmw br qrl; see already Peiser 1898: 6.
fact, a clear yod and nun in šʾyn—a perfectly intelligible word. 126 As in Katumuwa: ʿbd pnmw.
123 Lidzbarski’s he is not to be seen on the stone, nor in any of the 127 Lidzbarski (ESE 3: 198) thought he saw mem and taw in posi-
photos. tions 2–3, which could make one think of the Sam’alian particle mt

ters in positions 5–7 read bny. This word is already well linked to the beginning of line 4.134 Znn is already docu-
documented in Sam’alian as the plural of br, “son,” with mented as demonstrative pronoun in the Katumuwa in-
suffixed personal pronoun “my sons,” in Hadad 15, 20 scription, lines 8 and 9 (Pardee 2009: 64).
(mnmn bny), Panamuwa 5 (ḥd bny), and Katumuwa 6
(mn mn bny). Also in our inscription, this may refer to Line 4
sons and possibly successors of the deceased. A construct
state in the plural could alternatively have been meant, Joining the end of line 3 to the beginning of line 4
“the sons of. . . ,” as in Katumuwa 7 (mn bny ʾš) or a ver- renders lyhnyʾš. . . ,135 in turn possibly comparable to
bal form of bny, “build.” At the end of the line, ḥlbbh wzyhny•ʾšm• in line 10. One might understand lyhny
can clearly be made out,128 a Sam’alian word attested in as a precative of hny, with ʾš. . . (restored ʾš[m]) its di-
Hadad 3, 9, 10, 12, 13, 19 and translated “reign, authority, rect object,136 “may he do good to the name.”137 If this
sovereignty, succession,”129 according to the royal con- reading is accepted, it will have been followed by either
text—ḥṭr ḥlbbh.130 This word is preceded by nun-bet or a personal name or a suffixed personal pronoun of the
waw-bet. If the latter reading is accepted, wbḥlbbh may first or third person y or h, now much effaced. This in-
be related to the preceding blḥ . . . (less likely btḥ . . .), terpretation remains conjectural, however: the proposed
for example, blḥmh wbḥlbbh, “by fighting and by suc- nun can alternatively be read as pe, and ʾš understood as
cession,” or blḥwd, “alone.”131 Yet no such association is “man” (cf. Hadad 13, 34; Katumuwa 7), with yet more
known from the other Zincirli inscriptions. possibilities of restoration and explanation. For letters
10–13, the reading bḥrb, “by the sword,” may perhaps be
Line 3 proposed, an expression attested in Hadad 25 (just ḥrb
in Hadad 9 and Panamuwa 5). At the end of the line,
The first four(?) letters are illegible. There follows we suggest to restore wbny, “and my sons/descendants,”
probably sʿd in positions 6–8, with a possible dalet or he, though hesitatingly, as the reading and word division are
less likely yod, in position 5. Indeed, ysʿd is documented uncertain—the nun may be followed by a divider, and the
twice in the Hadad inscription (lines 15 and 21) in the word may continue in the next line.
expression wysʿd ʾbrw, “and he will affirm his power,”
with Lipiński (2000: 233, n. 6) proposing “he should Line 5
sustain” for the Ördekburnu sequence.132 The following
3–4 letters might be the direct object of ysʿd, but their The sense of the first four letters, even their exact
identification is too uncertain for a proposed reading. In identification, is not clear. The context could favor a
the following, ʿl•znnlyh is quite possibly legible.133 While qualification—an adjective or participle—bestowed by
no word divider is visible between the nun and lamed, the dynastic god, comparable to nʿm wḥny in line 6.138
we propose dividing ʿl•znn lyh, with the last three letters The identification of all following letters of line 5 has al-
ready been established by Lidzbarski (ESE 3: 200, 201),
and their reading is certain: . . . hʾ lrkbʾl ʾlh, “. . . he to/by
(Dion 1974: 179: “certainly”; Tropper 1993: 72: “indeed”), or an Ak-
Rakibʾil the god” (on ʾlh, see further line 6). The personal
kadian loanword, “country,” “land” (Lipiński 2000: 233, n. 6), or else a pronoun hʾ is probably the second element in a pronomi-
form of the verb mwt, “die”; yet the probable existence of a third letter nal proposition (Tropper 1993: 189)—see zkr hʾ in Hadad
before the following word divider renders this implausible. 30 and bʿl ksp hʾ in Panamuwa 11. It should be noted
128 Thus already Lidzbarski (ESE 3: 197, 201), further Lipiński 2000:
that Rakibʾil in Kulamuwa 16 and Panamuwa 22 is the
233, n. 6.
129 Possible derivation from ḥlp (> ḫlp), “replace, succeed,” and if so, dynastic god, bʿl b(y)t, and that in our inscription he is
comparable to Arabic ḫilāfa, “caliphate” (TSSI 2: 70), but in Sam’alian mentioned as first deity. In the inscription of Katumuwa,
with reduplicated third radical (Dion 1974: 88–89).
130 Sam’alian ḥṭr ḥlbbh may be compared to Akkadian ḫaṭṭi šarrūti, 134 For comparison, the Katumuwa inscription has three instances
“scepter of kingship,” Ugaritic ḫṭ mṯpṭ, Phoenician ḥṭr mšpṭ, “scepter of scriptio continua—in lines 7 and 8.
of rule,” and biblical Hebrew šbṭ mlkwt in Ps 45: 7 (Tawil 1974: 46–47; 135 Alternative reading lyhpyʾš. . . .
Fales 1982: 76–77). 136 The nun (or pe), yod, and alep are each followed by a potential
131 Cf. Nabataean blḥd (CIS II: 209, 7), blḥwd (CIS II: 215, 6). Or divider (see “Identification . . .” above). Obviously only one of them
eventually tḥwm, “territory.” The expression tḥtny, “in my place,” is real.
would fit the context of “succession,” but tḥt seems never to be preceded 137 Whereas Tropper (1993: 83) argued that in Sam’alian precative l
by the preposition b. before yqtl causes the yod to disappear, excepting a single uncertain oc-
132 The sequence ysʿd was one of the possibilities already consid- currence in Hadad 23, Katumuwa 8 has lw yqḥ (Pardee 2009: 64 right).
ered by Lidzbarski (ESE 3: 198). 138 One may be tempted to correct the possible whnp into wḥnp in
133 It seems to us that Lidzbarski’s zrmlyh cannot be maintained (see order to obtain a plausible “and the favor” (cf. Sokoloff 2002: 474), yet
“Identification . . . ” above). admittedly this would be rather strained.

a high official, he is not mentioned. On the face of it, this Kubaba and the moon (god). An alternative would be to
should indicate the royal nature of our inscription (more separate ʾrm from ly,144 understanding the former as a
in “Conclusions”). hitherto unknown (and admittedly uncertain) appella-
tion, “Kubaba of Aram,” and linking the latter to slʾ in
Line 6 the following line.
The reading of the first eight letters is somewhat Line 7

uncertain, though ?nʿm wḥny may suggest itself. If the
probable yod in position 1 is attached to the preceding The first three letters seem to read slʾ, a verb attested
word, we may understand ʾlhy as either “god,” a Sam’alian chiefly in Sabaean (s3lʾ) with the meaning of “make over,”
oblique plural of majesty with -y (cf. Tropper 1993: 74, “hand over” (Sabaic Dictionary: 138). The verb slʾ is doc-
200) or “my god,” with the personal pronoun -y.139 The umented also in Hebrew—in Lamentations 4.2, in an
pair nʿm wḥny could be understood as two qualifica- unpublished ostracon, and in personal names (the latter
tions related to the following personal pronoun hʾ, “he.” in Noth 1928: 174; WSS: 512), with the meaning of “pay”
Whereas it seems not to be attested in Aramaic, nor or the like. Preceded by ly if our above suggestion is ac-
hitherto in Sam’alian, nʿm might signify “good,” “agree- cepted, we have here a precative of the third person mas-
able,” as in Phoenician and Hebrew. The second word, culine singular lyslʾ, “let him make over, pay, present.”
ḥny (from ḥnn), is attested in the second Barrakib in- The following letters 4–8 are mostly uncertain. Aramaic
scription, line 8, and probably in Panamuwa 22–23: [(w) mnḥh, or rather mnḥ-, “offering,” might ensue, and even
ytn•r]kbʾl•ḥny, “and may Rakibʾil give me grace” (lit. “my if this is not the case, a similar sense could be suggested
grace”; Tropper 1993: 131, 143, 162, 164).140 At the same by the context.
time, it remains unclear to what the probable suffix -y The next letters are perfectly clear, and they have been
of the first person singular refers; this is why one could identified already by Peiser (1898: col. 7), Lidzbarski
think alternatively of a passive participle of the Sam’alian (ESE 3: 197), and Bossert (1932: 87). At the end of the
verb ḥnʾ/ḥny, as in Hadad 19 and Panamuwa 12, with the line, šʾ, “ram/sheep,” is to be read,145 a word likewise doc-
meaning “be allotted a place,” “be at rest.” umented in Katumuwa 9. This is evidently the masculine
The following mention of the goddess Kubaba has of šʾh, “ewe,” known from Panamuwa 6 and 9,146 and it
been recognized first by Bossert (1932: 35–36, n. 2). provides a parallel to ybl in Katumuwa 3–5.
Written kbbw, her name is found in Katumuwa 5 (Pardee
2009: 62),141 and written kbbh, in a fragment of a stone Line 8
inscription in relief from Tell Sifr (Tocci 1962) as well
as in the Aramaic inscription from Bahadirli in Cilicia The reading yn of the first two letters, followed by a
(KAI 278: 3, 7). Kubaba is amply documented as the word divider, is certain; they have already been recog-
great goddess of Carchemish (see, for instance, Hawkins nized by Lidzbarski (ESE 3: 199 [though he considered
1981; Radner 2005). In our inscription, her name is ym too]) and Bossert (1932: 87).147 Together with šʾ at
followed by the enigmatic five-letter ʾrmly. One would the end of the preceding line, they read šʾyn, “two sheep.”
expect an appellation or a toponym, but none of these Repeated twice, this is the first documentation of the
seems evident here.142 Meriggi (1975: 220) and Lipiński absolute dual ending in Sam’alian,148 though it is well
(2000: 234, n. 9) understood kbb ʾrmly as “lunar/monthly known in Old Aramaic (Segert 1975: 183–84, §5.2.3.3.5)
Kubaba,” deriving from Hittite Arma, “moon,” “moon
god,” though it was Landsberger who first proposed to 144 The lack of a word divider is probably no serious hindrance;
read the name of the god Arma here.143 At the same time, see n. 134.
Meriggi himself doubted the link thus created between 145 Lipiński 2000: 233, n. 6: “sheep.”
146 Notwithstanding Pardee’s doubts (2009: 65). Yet Pardee has
rightly noted that Ugaritic and Phoenician š (ś?) and Hebrew śh,
139 Less likely it belongs to the following word, making nʿm into “sheep,” are distinct from our šʾ. The latter is the same as Ugaritic *ṯa
an imperfect. (only feminine ṯat documented).
140 Cf. Phoenician ḥn lʿn ʾlnm, “favor in the eyes of the gods” 147 Lipiński (2000: 233, n. 6) read yn too, but his proposal to under-
(Yeḥawmilk, KAI 10, 10). stand this as “wine” is to be abandoned.

141 On the writing kbbw, see also Younger 2009: 166–79. 148 Tropper (1993: 201 with in-text n. 1) observed the absence of the
142 A link with Aramaic ʾrmlh, “widow,” evoked by Lidzbarski (ESE absolute dual from the Sam’alian texts known in his time, but has admi-
3: 205), seems to us excluded. rably foreseen its form: “Über die Form des Du. (Stat.abs.) im Samʾal.
143 1948: 49, n. 124: “. . . vollends Arma Z. 6.” Differently Lands- sind wir leider nicht unterrichtet. Im Hinblick auf den akkad. Befund
berger 1948: 64, n. 166: “Wortungetüme wie . . . ʾrm lysl erklären sich . . . liegt die Annahme nahe, daß das Samʾal. die Nunation nur im m.Pl.,
aus dem Fehlen des Worttrenners. . . .” nicht aber im Du. aufgegeben hat.”

and biblical Hebrew (Joüon 1965; Joüon and Muraoka also by the end of Katumuwa, where the deceased de-
2006: § 91). The understanding as a dual results also from mands of a future descendant that “he offers/roasts for
the contents: As a sacrifice is mentioned, the number of me a leg (of sheep)” (“. . . yšwy ly šq”; cf. Masson 2010:
sacrificed animals has to be precise, and in this context, 55–56; on “roast,” see n. 176 herein). Thus, Ördekburnu
šʾyn can mean nothing but “two sheep.” The reading of line 9 might be understood: “And in the necropolis of the
the next word, lym, “for the day,” is also certain. Our šʾyn kings two sheep for me” (see “Conclusions” on the royal
lym is close to šʾ ywmn lywmn in Katumuwa 9,149 “day,” in necropolis).
the latter being written with waw. Even though it is not
explicitly identified, “day” quite probably refers to the an- Line 10
nual celebration of the mortuary rites, the local parallel
to the Akkadian kispu (see “Conclusions”). In two minds as he was about his classifying the lan-
The remaining letters of line 8, except position 6, have guage of our inscription as non-Semitic, Lidzbarski (ESE
all been recognized by Bossert (1932: 87), many of them 3: 201) hazarded a West Semitic translation of line 10, his
already by Lidzbarski (ESE 3: 197). Bossert has also ob- only such continuous attempt, as a curse: “wer auslöschen
served that the expression here parallels the one in lines sollte den Namen des Ḫaian, sei verwünscht beim Gotte
7–8. We propose to read from the end of line 6 to the Šemeš,” at the same time not discarding the alternative
end of line 8 “Let him present an offering: for Rakibʾil of a blessing: “wer gedenken sollte des Namens des Ḫaian
two sheep for the Day and for Kubaba two sheep for the . . . sei gesegnet (oder ähnlich) beim Gotte Šemeš.”151 We
Day.” Obviously, the preposition b (in brkbʾl, bkbb) does share, for instance, Lidzbarski’s reading of ʾšm, “name,”
not stand for the usual locative here. It parallels the Katu- and the benedictory choice, also zy, “he who,” and thy,
muwa usage of l before the divine names and before nbšy “be,” but neither the reading kyn as a supposed rendering
in lines 3–5, and more directly of the b in wyhrg bnbšy in of the royal name Ḫaian nor the name of the sun god šmš.
line 11. This b may thus be characterized as directional, The beginning of the line, wzyhny•ʾšm, is clear and
signifying “for (the benefit of),” in particular after verbs may be understood as haplographic writing of wzy yhny
of motion or transfer such as our slʾ (Segert 1975: 349, ʾšm.152 The verb hny (also lines 3–4?) is hitherto known
6.5.2.2.3).150 in Judaeo-Aramaic (DNWSI I: 289) and in Syriac (DRS:
428–29), whereas the Sam’alian writing ʾšm with prothetic
Line 9 alep for šm, “name,” customary in Aramaic, is attested in
Hadad 16 and 21 as well as at Sfire (KAI 222C: 25; 223B:
In view of Lidzbarski’s numerous hesitations, the 7). The beginning of the line may be understood as “And
reading of line 9 might appear difficult at first sight. he who does good to the name.”153
Nevertheless, our collation finally yielded an interpre- Following the parallel in Hadad 21, the name of the
tation, which can perhaps be elucidated further by the deceased was probably contained in the next letters, posi-
end of the Katumuwa inscription. At the beginning of tions 10–13, which are hardly legible: the first two read
the line, wbmqm may be read, followed by a word di- perhaps py, though very tentatively,154 and if so, the name
vider, which befits our mortuary context: mqm signifies might have been Luwian beginning in Piya-, “gift, to
“(holy) place” in Sam’alian, more specifically “(resting) give,” a common element in personal names.155 We can-
place, tomb, necropolis” (Hadad 14; Tropper 1993: 75, not exclude the possibility that letters 14–16, probably
273, 279). The following mlky could be a Sam’alian plu-
ral, genitive—“of the kings” (cf. Tropper 1993: 199–200). 151 Lipiński (2000: 233, n. 6) proposed “. . . guilt-offering(?), when the
The identification of the rest of line 9 proved more dif- Sun-god has yielded the wine of this land.” To us, this is unacceptable.
ficult, yet combining the actual remains of the letters 152 Cf. the writing ʾlh•yʾdy for ʾlhy•yʾdy in Panamuwa 2 (Tropper
with the sense of the preceding lines, the tentative (and 1993: 176), where a word divider is present whereas in our inscription
we hope materially justified) reading emerged: šʾyn ly, it is absent. Hebrew has a comparable ḥyhwh for ḥy yhwh in Lachish
“two sheep for me.” To a certain extent, this is supported ostracon 3, line 9 (Aḥituv 2008: 68).
153 The alternative reading of ʾšm, i.e., ʾšr, “(resting) place” (see
“Identification . . .”), could also fit our present context. Indeed, “name”
149 See Pardee 2009: 65 on this expression; also Lemaire 2013: 149– and “place” are associated in a similar context in the first Nerab stele
50. Rather than a non-Sam’alian plural (Pardee 2009: 58, 65, 66, 67), (KAI 225), line 10, where “name” lacks the prothetic alep: šmk wʾšrk,
ywmn lywmn may perhaps be interpreted as a frozen singular form with “your name and your (resting) place.”
nunation, used in Sam’alian. 154 It is here that Lidzbarski opted for kyn; see “Identification. . . .”
150 Despite Pardee (2009: 54, 66), who preferred the locative mean- 155 E.g., pytrʿnz in Egypt (Kornfeld 1978: 116; TAD A6 7:4), srmpy
ing in Katumuwa 11, the directional alternative of this b appears like- in Cilicia (Lemaire 1993: 13), and Sandapî in cuneiform transcription
lier, all the more so as it is comparable to ly in Katumuwa 13, also in (R. Pruzsinszky in PNAE 3/I: 1087–88). See also Houwink Ten Cate
Ördekburnu 9 if our proposed reading is correct. 1961: 175–77.

thy, still belong to the name of the deceased.156 Yet these by the letters (see also n. 162); Barrakib is in between,
three letters may rather be understood as a third per- and Hadad and Panamuwa have no line separators. The
son feminine singular imperfect of hwh, “to be,” with the lamed, whose arm habitually rises above the rest of the
feminine subject nbš, “‘soul’, spirit of the dead,” whose letters in West Semitic inscriptions, is in our inscription
writing with bet is common at Zincirli (and elsewhere): constrained by the top line separator and is consequently
Hadad 17, 21, 22; Panamuwa 18; Katumuwa 5, 11; also often rather small (with the exception of letter 24 in line
Kulamuwa 13; Barrakib 2: 7 (ample literature, e.g., Fales 10, where there are no line separators). In this respect,
1982: 75–78). This interpretation may have a drawback, it follows Kulamuwa’s lameds. Katumuwa’s lameds are
however: it would be unusual that the only nbš men- larger, at times stretching down to the lower line sepa-
tioned in the text is not that of the deceased. Returning rator, and Barrakib’s are intermediate. The Hadad and
to hwh, at Sfire the Aramaic feminine form thwy is docu- Panamuwa lines are crowded together so that, despite the
mented (KAI 222A: 25, 32; 223A: 6), but Katumuwa 7 has absence of line separators, the lameds have only a little
masculine yhy, so that our thy may well be Sam’alian.157 It room to extend beyond the top of the line.158
is equally difficult to decide whether the possible he in The following comments are on letters that differ from
position 20 constitutes the suffixed personal pronoun of one Zincirli inscription to another—he, waw, zayin, ḥet,
nbš, “his/her ‘soul,’” as in Panamuwa 20, or not. The last kap, mem, pe and qop (see Figs. 20, 21).
letters of line 10 are even more uncertain, save the clear In some of the Ördekburnu hes, the shaft is convex,
penultimate lamed.158 in particular toward the bottom end. Among the other
In conclusion, we offer (see below) a continuous, Zincirli inscriptions, this is typical of Kulamuwa (Trop-
though incomplete transliteration and translation of this per 1993: 165).159
heavily damaged text while reminding the reader of the Two variants of waw seem to be present—the habitual
tentative or conjectural nature of certain parts. Ч-shape and the archaizing Y‑shape.
The zayin is of the archaizing, sideways-H type, as in
the Kulamuwa and Hadad inscriptions, eventually re-
The Writing placed in the inscriptions of Panamuwa II and Barrakib
by the more advanced Z‑shape. But that form is already
Palaeography present in the inscription on the bronzes of Hazael (Bron
and Lemaire 1989: 39), a contemporary of Kulamuwa.
Most letters in lines 1–9 (Fig. 18) hang from the top Not strictly speaking monumental, yet carved in a hard
line separator; line 10 (Fig. 19) is different. As a result, material, this zayin underlines the limitations of a pa-
the legless letters in the first nine lines—ʿayin, shin . . . , laeographic dating of monumental inscriptions as op-
but also others—have a conspicuous space underneath. posed to cursive inscriptions rapidly written in ink.
This is similar to Kulamuwa; in Katumuwa, the space be- The Ördekburnu ḥets are three-runged, as in all Zin
tween the separators is narrow and hence, as a rule, filled cirli inscriptions except Barrakib. They are also inclined
1 [ʾ]n[k]-----n-------------y--- I? [am PN]..........................

2 -----bny•blḥ(btḥ)---wbḥlbbh ...... my sons in ...... and in (the) kingship/succession
3 ----ysʿd•h---•ʿl•znn lyh ...... he will affirm ............ on this. Let him
4 ny•ʾš[m]--s---bḥrb•wbny do good for the na[me]...... by the sword. And my sons
5 --h--•hʾ•lrkbʾl•ʾlh .......... (be) he to Rakibʾil the/my god;
6 y nʿm•wḥny•hʾ•lkbb•ʾrm ly agreeable and gracious / well placed (be) he to Kubaba of Aram. Let him
7 slʾ mnḥ- brkbʾl•šʾ present an offering: for Rakibʾil two sheep
8 yn•lym•wbkbb•šʾyn•lym for the Day, and for Kubaba two sheep for the Day,
9 wbmqm•mlky•šʾyn•ly and in the royal necropolis (“[resting] place of the kings”) two sheep for me.
10 wzy <y>hny•ʾšm•py-- thy•nbšh---l- And he who will do good for the name of Piya-X let his ‘soul’ be ......
156 In that case, it may perhaps be compared to the woman’s name
ʾšlthy, written in Phoenician on a Cilician seal (Lemaire 1977: 33–38;

1991: 134–35). 158 It is tempting to restore thy nbšh lʿlm, “may his soul be forever!”
157 Pardee’s proposal (2009: 64) to correct Katumuwa’s yhy into yhwy at the end of the line, but such an interpretation is eminently conjectural.
is thus not necessarily called for. See further Pardee 2009: 68: “[T]he 159 “Der Zeichenschaft ist in K1 sehr lang, stark nach links geneigt
superficially obvious derivation is from HYY. Samalian shows HWY und meist in sich gebogen. Die Länge des Schaftes und seine Neigung
in the suffix conjugation, but the prefix conjugation is unattested.” This nach links nehmen chronologisch ab. Gebogene Schäfte lassen sich
cautious observation leaves more than one possibility open. nach K1 nicht mehr beobachten.”

Fig. 20. The letters arranged from alep to taw (Sass, after Figs. 18c and 19c).
Fig. 21. A sample of the letters compared with those of the other Zincirli inscriptions (Sass). Kulamuwa: after Sendschirli 4: 377. Second kap:
Kulamuwa gold ferrule after Berlin Museum photograph (cf. Sendschirli 5: 102). Ördekburnu: after Figs. 18c and 19c. Hadad: after HNSE: pl. 22.
Katumuwa: after Struble and Herrmann 2009: 19. Panamuwa: after Sendschirli 1: 81. Barrakib: after HNSE: pl. 24.
top left, as in Kulamuwa and Hadad; in Panamuwa (II) Among the kaps, one example only is advanced, its left-
and Barrakib, the ḥets are habitually vertical (Tropper hand finger attached to the center of the middle one. The
1993: 166).160 same form is known at Zincirli from two occurrences on
Kulamuwa’s gold ferrule. As in most Zincirli kaps down
160 “Die Neigung des Zeichens nach links nimmt chronologisch ab, to Barrakib, the three fingers of the other Ördekburnu
die Sprossen werden zunehmend waagerechter.” kaps are archaizing, springing from the same point. In

addition, four of the five kaps have convex legs, a phe- monuments—of Panamuwa II and Barrakib—just a few
nomenon common at Zincirli in the Kulamuwa and Ha- letters are more advanced. Looking at all Zincirli inscrip-
dad inscriptions but not afterward (Tropper 1993: 166).161 tions together, it may be reasonable to suggest that the
The mems’ heads are mostly oblique, echoing the still earliest monument of Sam’al was subsequently regarded
earlier vertical head. The same is true for most Zincirli in- as a standard to be followed for the letter forms—strictly
scriptions. The more advanced, horizontal-headed mem for the first two generations or so, and with the occa-
is to be found elsewhere in the West Semitic realm—in sional concession to “modernity” thereafter. The fact that
the Nerab steles (KAI 225–26), ca. 700 b.c., and already all these inscriptions are in relief (a rare exception in the
in the Brej stele (KAI 201), ca. 800 b.c. The horizontal- West Semitic realm in the ninth–eighth centuries), again
headed Katumuwa mems are probably not advanced; following Kulamuwa, may point in the same direction.164
rather, their heads were “leveled” for lack of space.162 Coming back to the Ördekburnu stele, all its letter
The Ördekburnu pe possibly comes in two, or even shapes that we were able to identify are paralleled in the
three variants, each of which may alternatively be read Kulamuwa inscription (thus already Naveh 1970: 8),165
as another letter, however. dated toward the end of the ninth century, and recur in the
Our qop matches the closed, archaic-looking examples Hadad inscription, dated toward the middle of the eighth
in the older Kulamuwa and Hadad inscriptions. In the century. It is, in particular, the zayin, ḥet, kap, mem, and
more recent inscriptions of Panamuwa and Barrakib, the qop in monumental West Semitic inscriptions that have
head began to open, an advanced trait. Both variants oc- undergone changes between the later ninth century and
cur in the Katumuwa stele, and it is also for this formal later eighth (Fig. 21). Ördekburnu’s sideways-H zayin,
reason that we placed the Katumuwa letters in Fig. 21 three-rung ḥet tilted top left, oblique-headed mem, and
between Hadad and Panamuwa.163 closed qop are among Zincirli’s more archaic shapes; the
The only Ördekburnu taw identified with reasonable kap alone, besides the ubiquitous conservative version
certainty is vertical, a rare variant that existed beside the retained at Sam’al until the end, displays once an evolved
top-right leaning one. At least one of Katumuwa’s taws, in form at Ördekburnu, but it too is already found at the
qnt, line 1, is possibly similar, if shorter due to the small beginning of the series, in Kulamuwa’s gold ferrule. An-
vertical space available. other Ördekburnu trait paralleled chiefly in Kulamuwa
and Hadad is the convexity of the he and kap’s legs.
Dating All considered, some six of our stele’s diagnostic let-
ter shapes are closest to those of Kulamuwa and Hadad
Inscriptions of four kings, three of them with direct and at the same time more archaic than their later Zin
Assyrian synchronisms (Tropper 1993: 10–19; Gilibert cirli counterparts. This concentration is probably large
2011: 14–17), make Sam’al ca. 820–730 b.c. one of the enough, the Zincirli picture consistent enough, to be
best-documented places in the Levant for monumen- deemed chronologically significant even in a monumen-
tal West Semitic palaeography. Yet the letter shapes in tal inscription, and if so, it would situate the Ördekburnu
monumental inscriptions are notoriously conservative. stele sometime between ca. 820 and 760 b.c.166
In most cases, only the presence of more evolved forms,
themselves somewhat belated expressions of the innova- 164 The Zincirli letters in relief are thought to be inspired by con-
tions occurring in the cursive, can be considered a reli- temporary Luwian monuments in this border region. The fact that the
able foundation for dating. alphabetic inscriptions of Sam’al continue to the end to be carved in
At Zincirli, Panamuwa I, half a century after Kulamuwa relief is all the more remarkable, considering that in the Luwian world,
the number of stone inscriptions in relief diminished gradually in favor
and possibly more, practically imitated Kulamuwa’s letter of incised inscriptions: “In the Late Period, relief and incised inscrip-
shapes in his own Hadad inscription, while in the next tions coexist from earliest to latest, though it does appear that the trend
was towards ever-increasing use of incised forms” (Hawkins 2000: 4).
161 “Die untere Hälfte des Schaftes weist in den älteren Inschriften 165 Lidzbarski (ESE 3: 196) observed only that “. . . alle Zeichen,
(K1 bis H) eine Biegung nach links auf, während der Schaft in den soweit sie deutlich zu erkennen sind, die älteste Form des Alphabetes
späteren Inschriften gerade verläuft.” aufweisen. Zain hat noch die form [sideways H] . . .”
162 There is hardly enough room for Katumuwa’s inscription on the 166 Certainly our stele could not be later than ca. 730, when the lan-
initially uninscribed stele, and thus the line separators are minimally guage of the alphabetic Zincirli inscriptions shifted to Aramaic under
spaced and most letters inevitably squat. Barrakib. Nor, being Sam’alian, could it be pre-Kulamuwa. Moreover,
163 What is more, the Panamuwa (II) inscription was carved soon as long as Kulamuwa remains the earliest example of alphabetic Zincirli
after the king’s death in 733/732 b.c. by his son and successor Bar- writing dated historically, we do not consider the possibility of a still
rakib, so that Katumuwa’s stele is certainly earlier: The living Panamuwa earlier setting for Ördekburnu. Lidzbarski, on the other hand, wanted
named in the stele may well have been the just-mentioned second king the stele to be older: The absence of Assyrian inspiration in the scene
of this name (Pardee 2009: 57), though Panamuwa I (presumably not led him (ESE 3: 196) to date the stele “spätestens in der ersten Hälfte
too early in his reign) also becomes a candidate in case he died just a des 8. Jahrh. v. Chr.” But he remarked further that this was a terminus
few years before, as is commonly believed. ad quem; the monument could be significantly earlier, even of the 10th

Orthography 3 m sg hʾ – lines 5, 6 (Tropper 1993: 189).

Suffixed to nouns and prepositions (Tropper 1993:
Considering the bad state of the text, this section is to 191):
be regarded as tentative. 1 sg after prep ly (Pardee 2009: 67)
Consonants 1 sg after m sg noun ʾlhy?
We have nothing to add to Tropper 1993: 169–70, nor 1 sg after m pl noun bny? ʾlhy?
attestations of distinct Sam’alian/Aramaic phonology 3 m sg after f sg noun nbšh?
such as rqy for original rḍy. Demonstrative pronoun: znn? (cf. Pardee 2009: 64, 67)
Relative pronoun: zy (Tropper 1993: 193)
Vowels and Diphthongs
Possible matres lectionis in final position include: Nouns
yod for /ī/: m pl obl noun mlky Masc ʾlh ʾšm *br [bny?] ym mlk mqm šʾ
implied in rel pron adjoining next word—zyhny < zy Fem ḥlbbh ḥrb? mnḥh? nbš
yhny GN ʾrm?
1 c sg pron suff after preposition ly? DN kbb rkbʾl
uncertain: bny? thy? yhny? ḥny? Sg abs: ʾšm ym ḥlbbh ḥrb? mnḥh? ʾlh?
he for /ā/: f ending ḥlbbh Sg cs: mqm (mlky) bny? ʾlhy?
he for /ê/: 3 m sg pron suff nbšh? Pl abs obl: mlky (Tropper 1993: 198–9)
Pl + pron: bny? ʾlhy?
Matres lectionis in median position: Dual abs: šʾyn, first documentation (n. 148).
None. But note ym not ywm: On comparison with Construct state: mqm mlky
ywm- in Panamuwa 9, 10, 18 (Tropper 1993: 188) Definite article: none, as expected (cf. Tropper 1993:
and in Katumuwa 10 (Pardee 2009: 65), it appears 193–94)
that the two forms coexisted in Sam’alian.167 Adjective/participle: nʿm? ḥny
Diphthongs: Verbs
yod for /ay/: m du nom noun šʾyn Suffix conjugation: bny? (cf. Tropper 1993: 69).
Prefix conjugation: thy, <y>hny (first documentation
The Language both) lyslʾ? ysʿd?
Indicative: wzy< y>hny, ysʿd?
Selected topics only are listed; the state of preserva- Jussive: thy
tion of the stele does not allow for an exhaustive analysis. Precative: lyhny ? lyslʾ ?
Particles (cf. Tropper 1993: 228–30)
Phonology
Precative particle l
lyhny ? lyslʾ ? Both are uncertain. If they are accepted,
/p/ > /b/ change: nbš (Tropper 1993: 180–81), ḥlbbh?
they would show a retention of the yod, as against the
(see n. 129).
five examples assembled by Tropper (1993: 83–84)
Prothesis: ʾšm (Tropper 1993: 185).
where the yod is contracted. But the reading of all
Retention of diphthong: šʾyn.
Tropper’s examples, except one, is likewise uncertain.
Contraction of diphthong(?): For ym, see “Orthogra-
Cf. also Pardee 2009: 64, 67, on lw.
phy” with n. 167.
Prepositions
Morphology b in, by, with: 2 bḥlbbh, 4 bḥrb ? 9 bmqm; for: 7 brkbʾl,
8 bkbb.
Personal Pronouns l to, for: 5 lrkbʾl, lkbb, 8 lym × 2, 9 ly
Independent: 1 c sg ʾnk is possibly to be restored in ʿl on: 3 ʿl ?
line 1.
Conjunction
century. He finally settled for the first half of the ninth with the (errone- w: 2 w(?)bḥlbbh, 4 wbny, 6 wḥny, 8 wbkbb, 9 wbmqm,
ous) identification in line 10 of the name of king Ḥayyā(n), a contem-
porary of Shalmaneser III (858–824), and perhaps already of his father
10 wzy
Assurnasirpal (883–859). Lidzbarski had a large following on this; see
n. 14, “Identification . . .” line 10, and n. 110. Syntax
167 “[T]he word for ‘day’ is written with and without {w} in Sama-
lian, and opinions vary on how to account for this fact (see Dion 1974: Verbal phrase: verb + direct-object
72–76; Tropper 1993: 186–87)” (Pardee 2009: 67). lyhny ʾšm ? lyslʾ mnḥ-? [lyslʾ] šʾyn, wzy <y>hny ʾšm

Nominal phrase: adjective/participle + indep pers Conclusions

pron
5 . . . hʾ, 6 nʿm wḥny hʾ Even with the name and title of the deceased ef-
faced, and despite all other limitations and incertitudes
of our attempted reading and translation, there are in-
Glossary sights to be gleaned from the scene, inscription, and
geographical-historical context combined.
ʾlh god 5 ʾlh or ʾlhy
ʾrm Aram 6 ʾrm ? A Mortuary Stele
ʾšm name 4, 10 ʾšm ?
ʾšr (resting) place 4, 10 ʾšr ? (if not The study of both the relief and the inscription indi-
ʾšm) cates that the stele is a mortuary one. The banquet scene
b in, by, with; for 2 bḥlbbh, 4 bḥrb ? with one or two persons at table has been linked to the
9 bmqm; “for”: mortuary cult long ago (e.g., Orthmann 1971: 376–77;
7 brkbʾl, 8 bkbb Bonatz 2000a: passim; 2000b: 191). This interpretation is
bny build 2 bny ? now clinched further by the newly discovered Katumuwa
bny sons, sing. br 2 bny ? 4 wbny stele (Struble and Herrmann 2009: 29–33). What we have
hʾ he 5, 6 hʾ learned from the inscription on the Ördekburnu stele
hwy/hyy be 10 thy ? goes further in the same direction, with the references in
hny do good, benefit 3–4 lyhny ? 10 yhny lines 7–9 to offerings similar to those of Katumuwa, the
w and 2 w( ?)bḥlbbh, allusion in line 10 to the “name” (or “[resting] place”?),
4 wbny, 6 wḥny, and the mention of a nbš, or ‘soul’ (Struble and Herrmann
8 wbkbb, 2009: 29–33); we cannot tell, however, whether the nbš is
9 wbmqm, 10 wzy that of the deceased or of a descendant (see “Translation
znn this (m. dem. pron.) 3 znn . . .”). In line 9, a royal necropolis is mentioned, where
zy [he] who (m. rel. pron.) 10 wzy offerings are to be made.
ḥlbbh power, authority 2 bḥlbbh
ḥnʾ be restful, find rest 6 wḥny ? A Queen’s Stele?
ḥny favor, grace 6 wḥny ?
ḥrb sword 4 bḥrb ? The term ḥlbbh, Sam’alian “kingship,” “succession,”
ym day 8 lym × 2 or the like, was hitherto known from royal monuments
kbb DN Kubaba 6 lkbb, wbkbb only, similarly the symbols at the top of the stele.
l to, for “to, for”: 5 lrkbʾl, Rakibʾil is mentioned as first deity in lines 5 and 7.
lkbb, 9 ly, As indicated in Kulamuwa 16 and Panamuwa 22, he is
8 lym × 2 the bʿl•bt / bʿl•byt, “lord of the (royal) house.”168 These
l precative particle 3–4 lyhny ? 6–7 lyslʾ ? references to the dynastic god as first deity also in the
mlk king 9 mlky contemporary Ördekburnu stele (more below on the dat-
mqm (resting) place 9 wbmqm ing) may imply that the deceased belonged to the royal
nbš ‘soul’ 10 nbšh family. The royal connection will be reinforced further if
nʿm agreeable 6 nʿm ? the following ʾlhy in lines 5–6 is to be understood as “my
slʾ give, present 6–7 lyslʾ ? god” (one of three possible readings).169
sʿd strengthen, support 3 sʿd/ysʿd ? The beginning of line 9 reads wbmqm mlky, “and in
ʿl on 3 ʿl ? the royal necropolis” (“[resting] place of the kings”). If
py. . . PN Piya-X 10 py. . . ? our reading of the end of the line is correct, it is offer-
rkbʾl DN Rakibʾil 5 lrkbʾl, 7 brkbʾl ings for the deceased that were to be made in the royal
šʾ ram/sheep 7–8, 8, 9 šʾyn (abs necropolis. At the same time, the reference to a “ne-
m dual) cropolis of the kings” does not necessarily indicate that
the stele commemorates a king. In fact, some details of
Specific Sam’alian Isoglosses
168 This role of the god may have changed later on: in his inscrip-
tions, King Barrakib refers to both Rakibʾil and the moon god bʿlḥrn
Mater lectionis -y denoting the abs m pl obl: mlky as mrʾy, “my lord” (Tropper 1993: 146).
(Pardee 2009: 58, 66). 169 It is thus perhaps no coincidence that Rakibʾil does not figure in
Vocabulary: ḥlbbh, znn? the non-royal stele of “Katumuwa, ‘servant’ of Panamuwa.”

the imagery, in particular the two spindles(?) held by the As Emilia Masson has justly observed, the funerary
deceased, are characteristic of the depiction of women rites described in the stele of Katumuwa (and, by exten-
(see “The Table Scene . . .”). One could thus think of a sion, in ours) are close to those of the Hittite Empire
queen, or queen mother.170 The person to her left is a period (for which see Hawkins 1980; 1989; Niehr 2001;
spouse or descendant, performing the mortuary rites. He 2004b: 406–10; Masson 2010: 54),175 and they may be
is certainly a man, as indicated by his mace—a typically compared to the Akkadian kispum (e.g., Tsukimoto 1985;
male marker and an indicator of rank, though in itself the Hallo 1992; Niehr 1994; 2004a; 2004b; 2006). It should be
weapon does not characterize him as royal. noted that the sacrifices mentioned in the Ördekburnu
Yet the royal character of the Ördekburnu stele, how- text are solely of sheep but that our deceased was to re-
ever plausible, cannot be regarded as definitely estab- ceive double the number anticipated in Katumuwa, prob-
lished: The number of inscribed non-royal monuments ably also in Hadad (18) (see Lemaire 2012).176
from the land of Sam’al available for comparison is not
enough; it comprises just the Karaburçlu and Katumuwa The Stele in Sam’alian Geography and History
steles.171 Moreover, the low-quality relief of the Ördek-
burnu stele stands in stark contrast to the exquisite unin- Where was mqm mlky, “the royal necropolis ([rest-
scribed mortuary stele, thought to depict a queen, from ing] place) of the kings,” mentioned in line 9? The site
Hilani I (see “The Table Scene . . .” with Fig. 16). If most of Gerçin (Fig. 7), 7 km northeast of Zincirli, seems to
of the above evidence seems to support the royal ances- have served as royal necropolis and mortuary cult place
try of the Ördekburnu deceased, the substandard relief for the kingdom of Sam’al, at least during the reign of
remains to be explained. Panamuwa I.177 The colossal statue of Hadad was found
there,178 and the mention in line 14 of mqm pnmw br
Mortuary Rites qrl mlk yʾdy may thus refer to Gerçin. It is tempting to
identify mqm mlky in Ördekburnu 9 with Gerçin. On the
The funerary or rather mortuary rites of the kings of other hand, the distance between the two sites exceeds 20
Sam’al have been the subject of numerous studies,172 the km, and one may alternatively seek an older(?) royal ne-
recent editio princeps and subsequent reexaminations of cropolis or mortuary cult place in the vicinity of Ördek-
the Katumuwa stele further enriching our knowledge in burnu itself.179 The two mounds and tumulus marked on
this particular domain.173 Winter and Luschan’s map (Fig. 5) a little to the north
It has often been remarked that the deceased was as- and northwest of Ördekburnu are of potential interest in
sociated with the deities and partook of the sacrifice. In this respect. Karapınar Höyük or Karapınar Mezarlık, a
the Ördekburnu inscription, this practice seems to be mound with an Ottoman-period cemetery, is the closest
mentioned in lines 7–9, with the offering of two sheep to among these to Ördekburnu, ca. 2 km, and probably the
the deceased after the same sacrifice has been performed likeliest source of our stele (Figs. 22, 23; arrow in Figs. 3
for Rakibʾil and Kubaba. The inscription points out that
the offering is to be made “for the Day,” obviously a well- 175 Cf. KUB 39.6 iii 18–24 (Otten 1958: 50–51): “. . . ils font des
defined procedure, celebrating either the anniversary offrandes comme suit: un mouton à la déesse Soleil de la Terre, un
mouton au dieu Soleil du Ciel, deux moutons aux Ancêtres (divinisés)
of the death or an annual feast on which mortuary rites mais un bœuf soit un bœuf d’engrais et un mouton d’engrais à l’âme du
were performed, possibly “The Day” par excellence, i.e., mort mais un mouton ils sacrifient au Jour Favorable.”
New Year.174 176 In Katumuwa 8–13, as we understand it (differently Pardee
2009: 65), a sheep is taken, then slaughtered for the “soul” of the de-
170 On the queen mother’s role in the Empire period, see Bin-Nun ceased, and a leg offered, or roasted for him (on “roast,” cf. Akkadian
1975; Melchert 2003: 18–20. Funerary rites for royal women are docu- šamû/šawû [CAD Š I: 350] and Arabic). The Katumuwa inscription also
mented already in the Empire period; cf. Otten 1958; Christmann- has ybl for sheep in lines 3, 4, and 5 (taken by Pardee to rule out the
Franck 1971: 61; Gonnet 1995: 193; Hout 1995: 199–201. Couples same, or similar, meaning for šʾ); likewise Panamuwa 21, where an-
too, royal or non-royal, are often represented in the region’s Iron Age other word, ʾmr bmšwt, probably “a lamb in roasted state” (hardly “in
monuments (e.g., Hawkins 1980: 216–18, pls. 6–7; Bonatz 2000a: 19– an offered state”), is employed besides. Apparently the authors had a
21, C26–28; 2000b: 190–91). precise idea in mind as they used the variants šʾ(h), ybl, and/or ʾmr in
171 Karaburçlu village is located 5 km north of Zincirli. For the stele the same inscription.
and its Luwian inscription, see Hawkins 2000: 276, included among the 177 Landsberger 1948: 66, n. 167; Tropper 1993: 4–5, 54; Niehr 1994:
monuments from Maraş in the unexplained absence of a Sam’al section. 68–71; 2006: 116–19; 2010: 280–81.
172 See, among others, Greenfield 1973; Tsukimoto 1985; Niehr 178 Likewise, the uninscribed fragments of three more colossi. The
1994; 2001; 2004a; 2004b; 2006. Panamuwa II statue, discovered at Tahtalı Pınar, midway between Zin
173 Struble and Herrmann 2009; Niehr 2010; Masson 2010: 54; cirli and Gerçin, was considered by the excavators to originate at the
Lemaire 2012: 133–36. latter site too (Sendschirli 1: 48).
174 Spronk 1986: 207 (though see the critique of Smith and Bloch- 179 That Ördekburnu, 18 km to the south of Zincirli, was still within
Smith 1988); Niehr 1994: 62; Lemaire 2012: 135–36. Sam’al’s borders seems to us as good as certain (see “Introduction”).

Fig. 22. Karapınar Höyük or Karapınar Mezarlık, 6 August 2011 (photo by Amir Fink).
Fig. 23. Looking from Karapınar Höyük south to Ördekburnu, 6 August 2011 (photo by Amir Fink).
and 5).180 This is the opinion of Amir Fink, who visited Ördekburnu (see “The Stele and Its Find-Spot” above).181
the area at our request on 6 August 2011. Fink remarked A promising site.
that the numerous Ottoman grave markers here could be In the absence of a royal name, the age of the Ördek-
reworked stones taken from the site itself and that our burnu stele can only be estimated. Our dating is founded
stele, possibly unearthed during the digging of a grave on the employment of Sam’alian, the palaeographic anal-
but, being unsuited for a tombstone, may instead have at- ysis, and to some extent the study of the imagery.
tracted the attention of the Kurdish felt-makers at nearby
181 Finkadded that Tartar Höyük, the second mound, is located
farther away from Ördekburnu, hence probably a less suitable candi-
date. The tumulus marked on Winter and Luschan’s map (Fig. 5) could
180 Mezarlık means “cemetery.” not be found.

• Sam’alian is documented in the inscriptions of Kula lines of the Ördekburnu inscription—ca. 182 letters in ca.
muwa to Barrakib. As Kulamuwa left a Phoenician in- 47–48 words—are therefore part count, part estimate.184
scription as well, and most of Barrakib’s inscriptions From the relatively clear lines 5–9, we inferred the ra-
are Aramaic, Sam’alian will have been in use from tio of letters to words in the rest of the front inscription as
some late point in the former king’s reign to early in a whole (the letter count in line 10 is as good as certain):
Barrakib’s reign, ca. 820–730 b.c. (Tropper 1993: 47;
Lemaire 2001: 188; Young 2002; Río Sánchez 2006: • Lines with 15–16 letters: ca. 4 words
178). • Lines with 17 letters (apparently none with 18): ca.
• As we have seen (“Palaeography and Dating”), the 4½ words
script seems to indicate a shorter time span, some 60 • Lines with 19–20 letters: ca. 5 words.
years, ca. 820–760 b.c.
• The study of the table scene probably yields a wider The number of words in lines 1–4, an estimate, ap-
margin and is thus of less use than the other two pears in square brackets in Table 1.
criteria. The 118 identified letters in Table 1 are the same as in
Fig. 20. Only the reasonably and uniquely identified let-
All the same, Bonatz (2000a: 21, 59, 68) hesitatingly ters, ca. two-thirds of the total, have made their way into
suggested the ninth century, possibly ca. 875–850. Genge Fig. 20. The remainder either have two or more potential
too (1979: 102) was for a dating before Kulamuwa, i.e., readings or are unidentified. Among the ca. 182 letters,
before ca. 830. However, these precise, ultra-high datings we follow Lidzbarski’s identification of ca. 140, at least in
may have been influenced by Lidzbarski’s attempt (ESE part;185 about half of the rest we read differently, and for
3: 200–201), followed by several other authors (most another half we have no identification.186
recently Tropper 1999: 231, 246; Brown 2008: 346), to
understand a conjectural sequence of the three letters Letter Frequency
kyn in line 10 as a rendering of the name of Ḥayyâ(n),
Kulamuwa’s father and a contemporary of the Assyrian That we have not recognized the rare letters gimel and
king Shalmaneser III, possibly also of his predecessor ṭet comes perhaps as no surprise.187 But that dalet, ʿayin,
Assurnasirpal II. Yet this understanding is to us plainly and taw are so infrequent at Ördekburnu may be a false
impossible—see n. 14, and “Identification. . . .” impression, created by the poor preservation of the stele.
All considered, we propose a dating between ca. 820 We are well aware that this limits the accuracy of our
and 760 b.c. for the Ördekburnu stele. If this is so, the proposed translation.
stele might belong to the time of Kulamuwa or a suc- In order to better assess the letter frequency, we com-
cessor, possibly qrl, the father of Panamuwa I, or else to pared the Ördekburnu inscription to the perfectly legible
an as yet undocumented predecessor. Very probably a Katumuwa inscription (Pardee 2009), whose only read-
woman, the deceased may have been a member of the ing problem is the similarity of dalet and resh.188 The re-
royal family. sult is presented in Table 2, arranged by the frequency
Appendix: How Many Letters and 184 Prepositions and conjunctions adjoining the following word
Words? Letter Frequency were not counted separately.
185 In part, where Lidzbarski had two alternatives of which we ad-
Word division in the Ördekburnu inscription, only mit only one, or when we point out two alternatives, of which only one
coincides with Lidzbarski’s choice. For line 8 and parts of lines 5–7, the
partly deciphered, is often in doubt; the state of preser- best preserved in the inscription, we agree also with Bossert 1932, aside
vation renders the recognition of word dividers difficult.182 from a single letter.
We hoped to find a clue in the well-preserved Katumuwa 186 For Lidzbarski, only four or five letters were so effaced that he
inscription with its 202 letters in 60 words, and 52 word refrained from proposing an identification, even a conjectural one, in
dividers.183 Yet, as a comparison of the letter frequencies both his text and transliteration (see “Identification of the Individual
Letters . . . ” and the charts in particular).
(below) shows, the Ördekburnu and Katumuwa inscrip- 187 Lidzbarski, too, found no gimel, yet he pointed out three ṭets in
tions may differ in this respect too. Our totals for the 10 lines 2 and 6, perceived as a dash within a circle. This is hard to accept:
such a form recurs in Barrakib’s inscriptions, whereas, following the
182 Thus already Lidzbarski: “Besonders schwierig war es, die Wort- Hadad inscription, we would expect the more archaic cross within a
trenner festzustellen” (ESE 3: 193). circle at Ördekburnu. Clear ṭets are missing altogether in our inscrip-
183 Among the nine Katumuwa words ending at the end of a line, tion, and the ones we pointed out in lines 4, 7, and 10 are admittedly
four (in lines 5, 6, 7, and 9) are followed by a word divider, and five (in doubtful.
lines 4, 8, 11, 12, and 13) are not. Besides, three mid-line words (in lines 188 This problem required that in Table 2 we count the dalet and
7 and 8) are not followed by a word divider. resh together in the Ördekburnu inscription too. Four letters—gimel,

Table 1. Word Count or Estimate Table 2. Ördekburnu–Katumuwa

Letter-Frequency Ratio
Line Letters Identified letters Words
1 ca. 19 2 [5] Ördekburnu Katumuwa Ratio
2 ca. 20 12 [5] alep 10.17% (12) 1.98% (4) × 5.14
3 ca. 19 11 [5] he 7.63% (9) 2.97% (6) × 2.57
4 17 8 [4½] bet 15.25% (18) 8.91% (18) × 1.71
5 15 12 4 (+ 1 letter) kap 4.24% (5) 2.48% (5) × 1.71
6 18–19 16 ca. 5 lamed 13.56% (16) 9.41% (19) × 1.44
7 ca. 16 10 3½–4 ḥet 2.54% (3) 1.98% (4) × 1.41
8 17 17 4½ samek 1.69% (2) 1.48% (3) × 1.14
9 15 15 4 shin 5.09% (6) 4.95% (10) × 1.03
10 25 15 6–7 yod 13.56% (16) 13.86% (28) × 0.98
Total ca. 182 118 ca. 47–48 mem 5.93% (7) 6.44% (13) × 0.92
nun 8.47% (10) 12.38% (25) × 0.68
zayin 1.69% (2) 2.97% (6) × 0.57
ratio—the Ördekburnu percentage divided by Katumu- ʿayin 0.85% (1) 1.48% (3) × 0.57
wa’s. For Ördekburnu, the percentage is calculated from dalet & resh 4.24% (5) 7.92% (16) × 0.54
qop 0.85% (1) 1.98% (4) × 0.43
the 118 reasonably identified letters in Fig. 20. On the
waw 3.39% (4) 11.39% (23) × 0.3
shading, see below. taw 0.85% (1) 2.97% (6) × 0.29
One should have thought that the letter frequency in gimel – 1.98% (4) –
two monumental inscriptions of the same (mortuary) ṣade – 1.48% (3) –
genre from the same land, written in the same language, pe – 0.99% (2) –
even though probably several decades apart, would be ṭet – – –
rather similar. Yet among the 17 letters available for Subtotal 100% (118)
comparison,189 the Ördekburnu frequency is signifi- Unidentified or ambiguous (ca. 64)
cantly different in 10 (shaded in Table 2)—lower than Total (ca. 182) 100% (202)
× 0.66 of the Katumuwa frequency or higher than × 1.5.
This could represent a bias, much of it no doubt due to
the high proportion, about one-third, of the uncertain or unidentified Ördekburnu letters, excluded from the
present exercise. But it cannot be ruled out that we erred
ṭet, pe, and ṣade—are not identified with certainty in the Ördekburnu also about some of the 118 letters characterized above as
inscription; ṭet is missing also in the Katumuwa inscription. “reasonably identified.”
189 See n. 188.
Acknowledgments
Dr. İsmail Karamut, Director of the Istanbul Archaeology Rodica Penchas and Yulia Gottlieb of the Institute of Archae-
Museums, granted us permission to study and photograph the ology, Tel Aviv University, who with skill and common sense
stele. In the museum on 21–22 April 2008, we were greatly have met the challenge of drawing this difficult inscription.
helped by Mrs. Zeynep Kızıltan, Deputy Director. Our sincere Special thanks are due to Amir Fink who, while excavating at
thanks to them. Ms. Evren Kıvançer of Istanbul accompanied Zincirli, visited the Ördekburnu area twice at our request, in
us in the museum, and her involvement contributed a great 2008 and 2011, and provided us with much valued information
deal to the success of the expedition. We are much indebted to and images.
References
Aḥituv, Sh. Bonatz, D.
2008 Echoes from the Past: Hebrew and Cognate In- 2000a Das syro-hethitische Grabdenkmal: Untersuchun-
scriptions from the Biblical Period. Jerusalem: gen zur Entstehung einer neuen Bildgattung in der
Carta. Eisenzeit im nordsyrisch-südostanatolischen Raum.
Bin-Nun, Sh. R. Mainz: von Zabern.
1975 The Tawanana in the Hittite Kingdom. Texte der 2000b Syro-Hittite Funerary Monuments: A Phenomenon
Hethiter 5. Heidelberg: Winter. of Tradition or Innovation? Pp. 189–210 in Essays

on Syria in the Iron Age, ed. G. Bunnens. Ancient ESE 3 = Lidzbarski, M.

Near Eastern Studies Supplement 7. Louvain: 1915 Die Stele von Ördek-burnu. Pp. 192–206 in Ephe
Peeters. meris für semitische Epigraphik, Vol. 3. Giessen:
Bossert, H. Th. Töpelmann.
1932 Šantaš und Kupapa: Neue Beiträge zur Entzifferung Fales, F. M.
der kretischen und hethitischen Bilderschrift. Mit- 1982 Note di semitico nordoccidentale. Vicino Oriente 5:
teilungen der altorientalischen Gesellschaft 6/3. 75–83.
Leipzig: Harrassowitz. Fitzmyer, J. A., and Kaufman, S. A.
1942 Altanatolien: Kunst und Handwerk in Kleinasien 1992 An Aramaic Bibliography, Part 1: Old, Official
von den Anfängen bis zum völligen Aufgehen in der and Biblical Aramaic. Baltimore: Johns Hopkins
griechischen Kultur. Ältesten Kulturen des Mittel- University.
meerkreises 2. Berlin: Wasmuth. Friedrich, J.
Bowman, R. A. 1932 Kleinasiatische Sprachdenkmäler. Kleine Texte für
1948 Arameans, Aramaic, and the Bible. Journal of Near Vorlesungen und Übungen 163. Berlin: de Gruyter.
Eastern Studies 7: 65–90. Genge, H.
Bron, F., and Lemaire, A. 1979 Nordsyrisch-südanatolische Reliefs: Eine archäo
1989 Les inscriptions araméennes de Hazaël. Revue logisch-historische Untersuchung, Datierung und
d’assyriologie et d’archéologie orientale 83: 35–44. Bestimmung. Historisk-filosofiske Meddelelser
Brown, B. 49/1–2. Copenhagen: Munksgaard.
2008 The Kilamuwa Relief: Ethnicity, Class and Power in Gilibert, A.
Iron Age North Syria. Pp. 339–55 in Proceedings of 2011 Syro-Hittite Monumental Art and the Archaeology
the 5th International Congress on the Archaeology of of Performance: The Stone Reliefs at Carchemish and
the Ancient Near East (5 ICAANE). Madrid, April Zincirli in the Earlier First Millennium bce. Topoi 2.
3–8 2006, Vol. 1, ed. J. M Córdoba, M. Molist, M. C. Berlin: de Gruyter.
Pérez, I. Rubio, and S. Martínez. Madrid: Univer- Gonnet, H.
sidad Autónoma de Madrid: Centro Superior de 1995 Le culte des ancêtres en Anatolie hittite au IIe mil-
Estudios sobre el Oriente Próximo y Egipto. lénaire avant notre ère. Anatolica 21: 189–95.
CAD = The Assyrian Dictionary of the Oriental Institute of the Greenfield, J. C.
University of Chicago. 1973 Un rite religieux araméen et ses parallèles. Revue
Çambel, H., and Özyar, A. biblique 80: 46–52.
2003 Karatepe—Aslantaş, Azatiwataya, Die Bildwerke. Hallo, W. W.
Mainz: von Zabern. 1992 Royal Ancestor Worship in the Biblical World. Pp.
Christmann-Franck, L. 381–401 in “Sha‘arei Talmon”: Studies in the Bible,
1971 Le rituel des funérailles royales hittites. Revue Hit- Qumran and the Ancient Near East Presented to
tite et Asianique 29: 61–111. Shemaryahu Talmon, ed. M. Fishbane and E. Tov.
CIS = Corpus Inscriptionum Semiticarum. Paris: Klincksieck. Winona Lake, IN: Eisenbrauns.
Cross, F. M., and Freedman, D. N. Hawkins, J. D.
1952 Early Hebrew Orthography: A Study of the Epi- 1980 Late Hittite Funerary Monuments. Pp. 213–26 in
graphic Evidence. American Oriental Series 36. New Death in Mesopotamia: Papers Read at the XXVIe
Haven: American Oriental Society. Rencontre Assyriologique Internationale, ed. B. Al-
Dhorme, P. P. ster. Mesopotamia 8. Copenhagen: Akademisk.
1930 Langues et écritures sémitiques. Études sémitiques 1. 1981 Kubaba. A. Philologisch. Reallexikon der Assyriolo-
Paris: Geuthner. gie 6: 257–61.
Dion, P. E. 1989 More Late Hittite Funerary Documents. Pp. 189–97
1974 La langue de Ya’udi. Ottawa: Editions Sr. in Anatolia and the Ancient Near East: Studies in
1997 Les Araméens à l’âge du Fer: histoire politique et Honor of Tahsin Özgüç, ed. K. Emre, B. Hrouda,
structures sociales. Études bibliques N.S. 34. Paris: M. J. Mellink, and N. Özgüc. Ankara: Türk Tarih
Gabalda. Kurumu.
DNWSI = Hoftijzer, J., and Jongeling, K. 2000 Corpus of Hieroglyphic Luwian Inscriptions, Vol. 1:
1995 Dictionary of the North-West Semitic Inscriptions. Inscriptions of the Iron Age. Untersuchungen zur
2 vols. Handbook of Oriental Studies 1, The Near indogermanischen Sprach- und Kulturwissenschaft
and Middle East 21. Leiden: Brill. N.F. 8. Berlin: de Gruyter.
DRS = Cohen, D.; Bron, F.; and Lonnet, A. HNSE = Lidzbarski, M.
1994 Dictionnaire des racines sémitiques ou attestées dans 1898 Handbuch der nordsemitischen Epigraphik nebst
les langues sémitiques, Fasc. 1. The Hague: Mouton. ausgewählten Inschriften. 2 vols. Weimar: Felber.
Duru, R. Hout, Th. van den
2004 Yesemek: The Largest Sculpture Workshop of the 1995 An Image of the Dead? Some Remarks on the
Ancient Near East. Istanbul: Türsab. Second Day of the Hittite Royal Funerary Ritual.

Pp. 195–211 in Atti del II Congresso internazionale Pp. 185–93 in La Cilicie: Espaces et pouvoirs locaux
di hittitologia, ed. O. Carruba, M. Giorgieri, and (2e millénaire av. J.-C.–4e siècle ap. J.-C.). Actes de la
C. Mora. Pavia: Iuculano. Table ronde internationale d’Istanbul, 2–5 novembre
Houwink Ten Cate, Ph. H. J. 1999, ed. E. Jean, A. M. Dinçol, and S. Durugönül.
1961 The Luwian Population Groups of Lycia and Cilicia Varia Anatolica 13. Paris: De Boccard.
Aspera during the Hellenistic Period. Documenta et 2012 Rites des vivants pour les morts dans le Royaume
Monumenta Orientis Antiqui 10. Leiden: Brill. de Sam’al (VIIIe siècle av. n. è.). Pp. 129–37 in Les
Joüon, P. vivants et leurs morts: Actes du colloque organisé par
1965 Grammaire de l’hébreu biblique. Rome: Institut bi- le Collège de France, Paris, les 14–15 avril 2010, ed.
blique pontifical. J.-M. Durand, Th. Römer, and J. Hutzli. Orbis Bibli-
Joüon, P., and Muraoka, T. cus et Orientalis 257. Fribourg: Academic.
2006 A Grammar of Biblical Hebrew. Rev. English ed. Sub- 2013 Le dialecte araméen de l’inscription de Kuttamuwa
sidia biblica 14. Rome: Pontificio Istituto b
iblico. (Zencirli, VIIIe s. av. n. è.). Pp. 145–50 in In the
KAI = Donner, H., and Röllig, W. Shadow of Bezalel: Aramaic, Biblical, and Ancient
1971 Kanaanäische und aramäische Inschriften. 3rd ed. 3 Near Eastern Studies in Honor of Bezalel Porten, ed.
vols. Wiesbaden: Harrassowitz. A. F. Botta. Culture and History of the Ancient Near
Kalaç, M. East 60. Leiden: Brill.
1975 Einige Stelen mit hieroglyphisch-luwischen In- Lipiński, E.
schriften oder ohne sie. Pp. 183–89 in Le temple 2000 The Aramaeans: Their Ancient History, Culture, Reli-
et le culte: Compte rendu de la vingtième Rencontre gion. Orientalia Lovaniensia Analecta 100. Leuven:
assyriologique internationale organisée à Leiden du Peeters.
3 au 7 juillet 1972. Publications de l’Institut histo- Masson, E.
rique et archéologique néerlandais à Stamboul 37. 2010 La stèle mortuaire de Kuttamuwa (Zincirli): com-
Leiden: Nederlands Historisch-Archeologisch In- ment l’appréhender. Semitica et classica 3: 47–58.
stituut te Istambul. Melchert, H. C.
Koopmans, J. J. 2003 Prehistory. Pp. 8–26 in The Luwians, ed. H. C.
1962 Aramäische Chrestomathie: Ausgewählte Texte Melchert. Handbook of Oriental Studies 1, The
(Inschriften, Ostraka und Papyri) bis zum 3. Jahr- Near and Middle East 68. Leiden: Brill.
hundert n. Chr. für das Studium der aramäischen Meriggi, P.
Sprache gesammelt. 2 vols. Leiden: Nederlands In- 1975 Manuale di eteo geroglifico, Vol. 2: Testi. Incunabula
stituut voor het Nabije Oosten. Graeca 13. Rome: Ateneo.
Kornfeld, W. Naveh, J.
1978 Onomastica Aramaica aus Ägypten. Sitzungs 1970 The Development of the Aramaic Script. Proceedings
berichte (Österreichische Akademie der Wissen- of the Israel Academy of Sciences and Humanities
schaften, Philosophisch-Historische Klasse) 333. 5/1. Jerusalem: Israel Academy of Sciences and
Vienna: Österreichische Akademie der Wissen- Humanities.
schaften. Niehr, H.
Landsberger, B. 1994 Zum Totenkult der Könige von Sam’al im 9. und 8.
1948 Sam’al: Studien zur Entdeckung der Ruinenstätte Jh. v. Chr. Studi epigrafici e linguistici 11: 57–73.
Karatepe. Veröffentlichungen der Türkischen his- 2001 Ein weiterer Aspekt zum Totenkult der Könige von
torischen Gesellschaft 7/16. Ankara: Türkische his- Sam’al. Studi epigrafici e linguistici 18: 83–97.
torische Gesellschaft. 2004a Götter und Kulte in Sam’al. Pp. 301–18 in Offizielle
Laroche, E. Religion, lokale Kulte und individuelle Religiosität:
1960 Les hiéroglyphes hittites, Vol. 1: L’écriture. Paris: Akten des religionsgeschichtlichen Symposiums
Centre national de la recherche scientifique. “Kleinasien und angrenzende Gebiete vom Beginn
Lemaire, A. des 2. bis zur Mitte des 1. Jahrtausends v. Chr.”
1977 Essai sur cinq sceaux phéniciens. Semitica 27: 29–40. (Bonn, 20.–22. Februar 2003), ed. M. Hutter and
1991 L’écriture phénicienne en Cilicie et la diffusion des D. Hutter-Braunsar. Alter Orient und Altes Testa-
écritures alphabétiques. Pp. 133–46 in Phoinikeia ment 318. Münster: Ugarit.
grammata: Lire et écrire en Méditerranée. Actes du 2004b Auswirkungen der späthethitischen Kultur auf
colloque de Liège, 15–18 novembre 1989, ed. C. Bau- die Religion der Aramäer in Südanatolien und
rain, C. Bonnet, and V. Krings. Studia Phoenicia 13. Nordsyrien. Pp. 405–44 in Die Außenwirkung des
Liège: Société des études classiques. späthethitischen Kulturraumes: Güteraustausch –
1993 Deux nouvelles inscriptions araméennes d’époque Kulturkontakt – Kulturtransfer. Akten der zweiten
perse en Cilicie orientale. Epigraphica Anatolica 21: Forschungstagung des Graduiertenkollegs „Ana-
9–14. tolien und seine Nachbarn“ der Eberhard-Karls-
2001 Les langues du royaume de Sam’al aux IXe–VIIIe s. Universität Tübingen (20. bis 22. November 2003),
av. J.-C. et leurs relations avec le royaume de Qué. ed. M. Novák, F. Prayon, and A. M. Wittke. Al-

ter Orient und Altes Testament 323. Münster: Sabaic Dictionary = Beeston, A. F. L; Müller, W. W.; Ghul,
Ugarit. M. A.; and Ryckmans, J.
2006 Bestattung und Ahnenkult in den Königshäusern 1982 Sabaic Dictionary. Louvain-la-Neuve: Peeters.
von Sam’al (Zincirli) und Guzāna (Tell Ḥalāf) in Sader, H.
Nordsyrien. Zeitschrift des deutschen Palästina- 1987 Les états araméens de Syrie depuis leur fonda-
Vereins 122: 111–39. tion jusqu’à leur transformation en provinces
2010 Bestattung und Totenkult. Pp. 279–84 in Religionen assyriennes. Beiruter Texte und Studien 36. Bei-
in der Umwelt des Alten Testaments II: Phönizier, rut: Orient-Institut der Deutschen Morgenlän-
Punier, Aramäer, by C. Bonnet and H. Niehr. Kohl- dischen Gesellschaft.
hammer Studienbücher Theologie 4/2. Stuttgart: Sass, B.
Kohlhammer. 2007 From Maraş and Zincirli to es-Sawdāʾ: The Syro-
Noth, M. Hittite Roots of the South Arabian Table Scene. Pp.
1928 Die israelitischen Personennamen im Rahmen der 293–320 in Images as Sources. Bilder als Quellen:
Gemeinsemitischen Namengebung. Beiträge zur Studies on Ancient Near Eastern Artefacts and the
Wissenschaft vom Alten und Neuen Testament Bible Inspired by the Work of Othmar Keel, ed.
3/10. Stuttgart: Kohlhammer. S. Bickel, S. Schroer, R. Schurte, and Chr. Ueh-
Orthmann, W. linger. Orbis Biblicus et Orientalis, Sonderband.
1971 Untersuchungen zur späthethitischen Kunst. Saar Fribourg: Academic.
brücker Beiträge zur Altertumskunde 8. Bonn: Sayce, A. H.
Habelt. 1927 The Moscho-Hittite Inscriptions. Journal of the
Otten, H. Royal Asiatic Society 59: 699–715.
1958 Hethitische Totenrituale. Deutsche Akademie der Sebbane, M.
Wissenschaften zu Berlin, Institut für Orientfor 2009 The Mace in Israel and the Ancient Near East from
schung, Veröffentlichung 37. Berlin: Akademie. the Ninth Millennium to the First. Typology and
Pardee, D. Chronology, Technology, Military and Ceremonial
2009 A New Aramaic Inscription from Zincirli. Bulletin Use, Regional Interconnections. Ph.D. dissertation,
of the American Schools of Oriental Research 356: Tel Aviv University (Hebrew).
51–71. Segert, S.
Peiser, F. E. 1975 Altaramäische Grammatik. Leipzig: Enzyklopädie.
1898 Aus dem Kaiserlich ottomanischen Museum in Sendschirli 1 = Luschan, F. von; Schrader, E.; and Sachau, E.
Constantinopel. Orientalistische Literatur-Zeitung 1893 Ausgrabungen in Sendschirli, Vol. 1: Einleitung und
1: cols. 6–9. Inschriften. Mittheilungen aus den orientalischen
PNAE = Baker, H. D., ed. Sammlungen 11. Berlin: Spemann.
2002 The Prosopography of the Neo-Assyrian Empire, Vol. Sendschirli 3 = Luschan, F. von
3/I: P–Ṣ. Helsinki: The Neo-Assyrian Text Corpus 1902 Ausgrabungen in Sendschirli, Vol. 3: Thorsculpturen.
Project. Mittheilungen aus den orientalischen Sammlungen
Przeworski, S. 13. Berlin: Reimer.
1928 Vier nordsyrisch-hethitische Denkmäler. Oriental- Sendschirli 4 = Luschan, F. von, and Jacoby, G.
istische Literatur-Zeitung 31: cols. 233–38. 1911 Ausgrabungen in Sendschirli, Vol. 4. Mittheilungen
Radner, K. aus den orientalischen Sammlungen 14. Berlin:
2005 Kubaba und die Fische: Bemerkungen zur Herrin Reimer.
von Karkemiš. Pp. 543–56 in Von Sumer bis Homer: Sendschirli 5 = Luschan, F. von, and Andrae, W.
Festschrift für Manfred Schretter zum 60. Geburtstag 1943 Ausgrabungen in Sendschirli, Vol. 5: Die Klein-
am 25. Februar 2004, ed. R. Rollinger. Alter Orient funde von Sendschirli. Mitteilungen aus den orien-
und Altes Testament 325. Münster: Ugarit-Verlag. talischen Sammlungen 15. Berlin: de Gruyter.
RES 3 = Chabot, J.-B. Smith, M. S., and Bloch-Smith, E. M.
1916– Répertoire d’épigraphie sémitique, Vol. 3. Paris: 1988 Review of Beatific Afterlife in Ancient Israel and in
1918 Klincksieck. the Ancient Near East, by K. Spronk. Journal of the
Río Sánchez, F. del American Oriental Society 108: 278–79.
2006 Diglosia en Arameo Antiguo. Pp. 173–81 in Šapal Sokoloff, M.
tibnim mû illakū: Studies Presented to Joaquín 2002 A Dictionary of Jewish Babylonian Aramaic of the
Sanmartín on the Occasion of His 65th Birthday, ed. Talmudic and Geonic Periods. Dictionaries of Tal-
G. del Olmo Lete, L. Feliu, and A. M. Albà. Aula mud, Midrash, and Targum 3. Ramat Gan: Bar Ilan
Orientalis Supplement 22. Barcelona: AUSA. University.
Rosenkranz, B. Spronk, K.
1978 Vergleichende Untersuchungen der altanatolischen 1986 Beatific Afterlife in Ancient Israel and in the Ancient
Sprachen. Trends in Linguistics. State-of-the-Art Near East. Alter Orient und Altes Testament 219.
Reports 8. The Hague: Mouton. Kevelaer: Butzon & Bercker.

Struble, E. J., and Herrmann, V. R. Uehlinger, Chr.

2009 An Eternal Feast at Sam’al: The New Iron Age Mor- 2003 Clio in a World of Pictures—Another Look at the
tuary Stele from Zincirli in Context. Bulletin of the Lachish Reliefs from Sennacherib’s Southwest Palace
American Schools of Oriental Research 356: 15–49. at Nineveh. Pp. 221–305 in “Like a Bird in a Cage”:
TAD A = Porten, B., and Yardeni, A. The Invasion of Sennacherib in 701 b.c., ed. L. L.
1986 Textbook of Aramaic Documents from Ancient Grabbe. Journal for the Study of the Old Testament
Egypt, Vol. 1: Letters. Jerusalem: Hebrew University, Supplement 363; European Seminar in Historical
Department of the History of the Jewish People. Methodology 4. Sheffield: Sheffield Academic.
Tawil, H. Voos, J.
1974 Some Literary Elements in the Opening Sections of 1986 Studien zur Rolle von Statuen und Reliefs im sy-
the Hadad, Zākir, and the Nērab II Inscriptions in rohethitischen Totenkult während der frühen
the Light of East and West Semitic Royal Inscrip- Eisenzeit (ca. 10.–7. Jh. v.u.Z.). Ph.D. dissertation,
tions. Orientalia 43: 40–65. Zentralinstitut für Alte Geschichte und Archäolo-
Tocci, F. M. gie, Akademie der Wissenschaften der DDR, Berlin.
1962 Un frammento di stele aramaica da Tell Sifr. Oriens Wartke, R.-B.
antiquus 1: 21–22. 2005 Sam’al: Ein aramäischer Stadtstaat des 10. bis 8. Jhs.
Tropper, J. v. Chr. und die Geschichte seiner Erforschung. Mainz:
1993 Die Inschriften von Zincirli. Abhandlungen zur Lit- von Zabern.
eratur Alt-Syrien-Palästinas 6. Münster: Ugarit. WSS = Avigad, N., and Sass, B.
1999 Sam’al – Schnittstelle der Kulturen. Pp. 225–47 in 1997 Corpus of West Semitic Stamp Seals. Jerusalem:
Von Babylon bis Jerusalem: Die Welt der altorien- Israel Academy of Sciences and Humanities.
talischen Königsstädte, Reiss-Museum Mannheim, Yadin, Y.
Kunsthistorisches Museum Wien. Bible Lands Mu- 1970 Symbols of Deities at Zinjirli, Carthage and Hazor.
seum Jerusalem, ed. W. Seipel and A. Wieczorek. Pp. 199–231 in Near Eastern Archaeology in the
Mannheim: Reiss-Museum. Twentieth Century: Essays in Honor of Nelson
TSSI = Gibson, J. C. L. Glueck, ed. J. A. Sanders. Garden City, NY: Double-
1975 Textbook of Syrian Semitic Inscriptions, Vol. 2: Ara- day.
maic Inscriptions Including Inscriptions in the Dia- Young, I.
lect of Zenjirli. Oxford: Clarendon. 2002 The Languages of Ancient Sam’al. Maarav 8: 93–
Tsukimoto, A. 105.
1985 Untersuchungen zur Totenpflege (kispum) im alten Younger, K. L.
Mesopotamien. Alter Orient und Altes Testament 2009 Two Epigraphic Notes on the New Katumuwa In-
216. Kevelaer: Butzon & Bercker. scription from Zincirli. Maarav 16: 159–79.

BASOR
Bulletin of the American Schools of Oriental Research • Number 369 • May 2013
Bulletin of the
American Schools of
Oriental Research
Number 369 — May 2013


Lemaire A. and Sass B. The Mortuar PDF

Uploaded by

Copyright:

Available Formats

Lemaire A. and Sass B. The Mortuar PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lemaire A. and Sass B. The Mortuar PDF

Uploaded by

Copyright:

Available Formats

The Mortuary Stele with Sam’alian Inscription

from Ördekburnu near Zincirli

André Lemaire and Benjamin Sass

Introduction The stele was pointed out to von Luschan at Ördek-

line had not yet been constructed.

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

While the course of Sam’al’s southern border is today

Attempts at a decipherment by Peiser (1898) and von

9 Von Luschan reported that “Unabhängig von Peiser habe ich

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 3. View of Ördekburnu (Google Earth, 22 April 2011, © Google).

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

gramme in hethitischer Hieroglyphenschrift. Unter dem Relief eine

the language of the text. identifier la langue.”

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

All use subject to JSTOR Terms and Conditions

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 6. Koldewey’s map (Sendschirli 1, at end).

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 7. Map by Steven D. Batiuk, 2012.

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 8. Lidzbarski’s drawing of line 10 (ESE 3: 194).

Fig. 9. Bossert’s partial drawing of lines 5–8 (1932: 36, 87).

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Two probably beardless figures face one another over a

32 Comparable spindles are to be found, for instance, in the hands

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 11 (continued). (b) front, inscription: unnumbered.

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 11 (continued). (c) front, inscription: Istan. 2072.

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

The mortuary character of the stele was already rec-

35 The stele is discussed with the “Grabstelen” in Sendschirli 4, and

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 18 (a–c). Front inscription: (a) photo 2901.

Line 1 Positions 3–7. We estimate five obliterated letters be-

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 18 (continued). (b) drawing superimposed on photo.

a dot attached to its right—hence a partly obliterated taw(?)—while dashes.

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

‫י א‬ ‫ה‬ ‫ע מ‬/‫ל‬ ‫ר‬/‫ב ד‬ ‫– י‬ • ‫נ‬ • ‫• כ‬/– ‫ש‬/‫א‬ ‫כ‬ ‫נ‬ ‫א‬

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Line 2 • Position 6. The nun is certain, even though its top

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

‫ב ח ל ב ב ה‬ ‫נ‬ •/‫ת‬ ‫א‬ ‫ת ח ש‬/‫מ ל‬/‫ט‬/‫• ב‬ ‫נ י‬ ‫ט‬/‫ב‬ ‫י‬ ‫ת‬ ‫מ‬ ?

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

This content downloaded from 132.66.7.210 on Thu, 25 Jul 2013 06:03:21 AM

Fig. 3. View of Ördekburnu (Google Earth, 22 April 2011, © Google).

Fig. 6. Koldewey’s map (Sendschirli 1, at end).

Fig. 7. Map by Steven D. Batiuk, 2012.

Fig. 8. Lidzbarski’s drawing of line 10 (ESE 3: 194).

Fig. 9. Bossert’s partial drawing of lines 5–8 (1932: 36, 87).

Fig. 18 (a–c). Front inscription: (a) photo 2901.

Fig. 18 (continued). (b) drawing superimposed on photo.