Везде
An Unpublished North West Semitic Inscription from the Collection of the Russian Archaeological Institute in Constantinople
Оглавление
Аннотация Оценить Содержание публикации
Библиография Комментарии
Поделиться
QR
Метрика
An Unpublished North West Semitic Inscription from the Collection of the Russian Archaeological Institute in Constantinople
Аннотация
Название публикации (др.)
Неопубликованная северозападносемитская надпись из коллекции Русского археологического института в Константинополе
Код статьи
S032103910023595-7-1
Тип публикации
Статья
Статус публикации
Опубликовано
Авторы
Балахванцев Арчил Савелич 
ORCID: 0000-0002-3028-0109
Аффилиация: Институт востоковедения РАН
Адрес: Российская Федерация, Москва
Коган Леонид Ефимович
Аффилиация: Национальный исследовательский университет Высшая школа экономики
Адрес: Российская Федерация, Москва
Лявданский Алексей К.
Аффилиация: Национальный исследовательский университет Высшая школа экономики
Адрес: Российская Федерация, Москва
Тункина Ирина Владимировна
ORCID: 0000-0001-5793-3977
Аффилиация: Санкт-Петербургский филиал Архива Российской академии наук
Адрес: Российская Федерация, Санкт-Петербург
Выпуск
Том 82 Выпуск 4
Страницы
911-940
Аннотация

The authors publish a hitherto unknown squeeze made from a North West Semitic inscription. The squeeze originally belonged to the Russian Archeological Institute in Constantinople and is now hosted by the St. Petersburg Branch of the Archive of the Russian Academy of Sciences. The inscription, performed in relief, may have been made on a large ornamented stone vessel. Paleographic features of the letters point to the mid-9th century BC as its approximate date. Both paleographic and linguistic features of the text suggest that its origin is to be sought in Southern Canaan (Palestine).

Ключевые слова
North West Semitic epigraphy, Palestine, Canaan, Semitic languages, squeezes (estampages), Russian Archaeological Institute in Constantinople
Источник финансирования
The study was conducted with the financial support of the Ministry of Education and Science of the Russian Federation, project no. 075-15-2020-786 “History of Writing of the European Civilization” («История письма европейской цивилизации»).
Классификатор
Получено
12.05.2022
Дата публикации
23.12.2022
Всего подписок
3
Всего просмотров
60
Оценка читателей
0.0 (0 голосов)
Цитировать   Скачать pdf Скачать JATS
1 1. The collection of estampages of the Russian Archaeological Institute in Constantinople
2 The St. Petersburg Branch of the Archive of the Russian Academy of Sciences hosts the collection of the Russian Archaeological Institute in Constantinople, hereafter RAIС (SPbB ARAS / SPbF ARAN. F. 127. Op. 1–3. D. 253. 1889–1925).
3 The RAIC was the only Russian scholarly institution operating outside the borders of the Russian Empire. The permanent director and driving force of the Institute was the distinguished Byzantinist Fyodor Ivanovich Uspensky (1845–1928). The Institute was engaged in archaeological and historical research in Greece, the Balkans, the Middle East, Asia Minor and other regions that once belonged to the Byzantine Empire. The RAIC was founded in 1894 and opened in February 1895. For almost twenty years, the Institute conducted systematic historical, philological and archaeological research within the borders of the Ottoman Empire, until the latter entered the First World War and broke off diplomatic relations with Russia. The Institute was closed in October 1914. The RAIC published its Proceedings (“Izvestiya Russkogo arkheologicheskogo instituta v Konstantinopole”): 16 sizable volumes appeared, the 17th was in preparation, but has never been published.
4 In the course of the RAIC’s archaeological expeditions and research missions, the Institute’s employees were able to study a number of monuments and art objects, as well as a few ancient and medieval inscriptions found during regular excavations or discovered accidentally. Estampages of inscriptions were often made in situ. It was an inexpensive and technically simple way to obtain exact copies of ancient texts with merely a brush and some paper1. Finds and gifts from private individuals entered the RAIC’s Cabinet of Antiquities, which soon grew into a small museum with a first-rate collection of monuments.
1. Rag vergé paper or blotting paper was used.
5 In October 1914, when the activities of the Institute were interrupted, only part of its archive could be evacuated to Russia. It was stored in Odessa, in the premises of the Historical and Philological Society of the Novorossia (Odessa) University. In 1920, the papers and books were moved to the Rare Books Department of the Central Scientific Library in Odessa, without any deed of transfer or inventory. In 1926, the documents were transported to Leningrad and entered the Byzantine Commission of the Academy of Sciences of the USSR.
6 For many years, Uspensky struggled for the return of the RAIC’s remaining scientific materials to Russia. However, this became possible only after his death, in 1929. The library, manuscripts, documents, prints, clichés, photographs and negatives were handed over by Turkey without inventories, and it may well be that some of the Institute’s property was actually not returned to the USSR.
7 A special commission created by the Academy of Sciences distributed the RAIС holdings among several institutions: the collection of manuscripts and books entered the Library of the Soviet Academy of Sciences and the Institute of History; the museum exhibits (94 boxes) were transferred to the State Hermitage Museum. Archival materials and paper squeezes were scattered between the Archive of the Soviet Academy of Sciences, the Leningrad Institute of History, Philosophy and Linguistics, the Leningrad Branch of the Central Historical Archive (currently the Russian State Historical Archive) and the Archive of the Revolution and Foreign Policy of Russia (currently the Foreign Policy Archive of the Russian Empire at the Ministry of Foreign Affairs of the Russian Federation)2.
2. More details can be found in Basargina 1995 and Tunkina et al. 2020.
8 Within the RAIC collection in the St. Petersburg Branch of the Archive of the RAS, the “separate inventory 3” represents the collection of estampages of the Russian Archaeological Institute in Constantinople (1895–1914), containing 51 items. Due to the general decline of Byzantine studies in early post-revolutionary Russia, the estampages lost their data sheets and, for a long time, were kept scattered in the SPbB ARAS.
9 2. General information about the estampage
10 In 2016, the epigraphist N.A. Pavlichenko made a primary attribution of the documents and established their origin3. She also compiled an inventory of the collection, in which the present inscription (SPbB ARAS / SPbF ARAN. F. 127. Op. 3. D. 51. L. 1) was mistakenly labelled “Arabic.” In late 2020, I.V. Tunkina examined the estampage, questioned the original identification and invited a trio of specialists in North West Semitic philology for a deeper inquiry into the paleography and contents of the inscription.
3. Pavlichenko 2018.
11 The following are the principal material characteristics of the piece4.
4. Described by N.V. Chernova, senior researcher at the SPbB ARAS.
12 The two-layer estampage was squeezed on four narrow strips of gray rag vergé paper (fig. 1). The state of preservation can be evaluated as very good (ca. 95%): the only disturbing features are some foxing on the paper and a tearing at the bottom. In the course of restoration, the estampage was unfolded.
13 The thin, long-staple cotton paper (0.110–0.126 mm thick) is plastic and hygroscopic, of the “blotter” type. The distance between the pontuseaux is 28 mm, the number of vergeures is 8 by 10 mm. The paper is, most likely, of European (rather than Asian) production of the late 19th – early 20th century. The estampage consists of two strips of two-layer paper, overlapped with a drop of glue (fig. 2). In the strips of the two-layer paper, the pontuseaux coincided: the pontuseau is located horizontally in one strip and vertically, in the other. That is, differently oriented pieces of paper were chosen for the two strips that make up the estampage. In those places where letters are located vergeures and pontuseaux are not violated. The overlap dimensions are: top – 160 mm, bottom – 189 mm. The overall dimensions are: length – 746 mm, width along the edges – 90 mm, in the central part – 70 mm. The letters are about 32 mm high. Most likely, a Greek sponge was used to make the squeeze: with it, wet rag paper was gently pressed into the surface of the inscribed stone.
14 Since the inscription consisted of just one line, only a narrow strip of paper was used. The upper part of the strip was held together with glue, then its lower edges were chipped off with a metal tailor’s pin to form a ring (fig. 3). The dimensions of the first strip: length – 585 mm, width – 90–70 mm; the dimensions of the second strip: length – 495 mm, width – 90–72 mm. The text is arranged on a raised stripe, the edges of which are clearly visible on the squeeze. The inscription might have run along the wide flat edge of a hemispherical or conical stone bowl5, or along the wide throat of a vessel, with shoulders ornamented with wide depressed rings. The upper parts of the rings can be seen on a 3D model of the estampage (fig. 4).
5. Cf. the well-known inscribed stone basin from Kuntillet ʕAjrud (Meshel 2012, 352, fig. 14.2).
15 The origin of the estampage cannot be established now due to a temporary closure of the archive, which is currently being transferred to a new building. There is hope that a careful perusal of the RAIC’s documents will sooner or later reveal the name of the person who made the estampage and the place where the object was found. It could be the director of the Institute, Uspensky himself, who had repeatedly visited Palestine, or one of his employees, or an undergraduate of the Russian universities and theological academies affiliated to the RAIC (notably, Ya.I. Smirnov, E.M. Pridik, B.V. Farmakovsky, M.I. Rostovtzeff, B.A. Panchenko, A.A. Vasiliev), or even a Russian diplomat. It is known, for example, that the Russian Consul General in Jerusalem, S.V. Arsenyev, was keenly interested in Palestinian archeology and repeatedly donated archeological monuments (including inscriptions) to the RAIC’s museum6.
6. See the 1895 report of the Russian Archaeological Institute in Constantinople, published in the Proceedings of the RAIC [Izvestiya RAIK]. 1896. T. 1, p. 40 (Отчет о деятельности Русского археологического института в Константинополе за 1895 г. Известия РАИК. 1986. Т. 1. Хроника, 23–53).
16 The most conspicuous technical feature of the inscription is the fact that it is performed in relief (Reliefschrift, champlevé). Among NWS alphabetic inscriptions, this technique is known almost exclusively from the Phoenician and Aramaic inscriptions found in Zincirli and its environs (9–8 c. BC)7, where the relief alphabetic script flourished under the influence of the Anatolian hieroglyphic script8. However, as one can deduct from the existence of a Phoenician inscribed stone fragment from Byblos (dated to the 10th century by its publisher, P. Bordreuil), this epigraphic style was also known in Phoenicia9. Later on, the relief script is attested by Aramaic and Dadanite inscriptions from Tayma10.
7. Pardee 2009, 52, fig. 1; Lemaire, Sass 2013, 57, fig. 2.

8. Osborne 2020, 142. Hereafter, all the dates in this article are BC.

9. Bordreuil 1977, pl. V.

10. Roche 2020, 171–191, fig. 1–10; Al-Ghabban et al. 2010, 255, 284–285, no. 103, 119, 120. Historical considerations led Roche to suppose that the relief script was brought to Tayma from Northern Syria during Nabonidus’ reign in Babylonia (556–539), in spite of the obvious chronological gap.
17 3. Paleography
18 3.1. General features (fig. 5, 6)
19 The text was inscribed in scriptio continua: neither spaces between words nor dividers are present. This is rather atypical for early Levantine epigraphs, in which word-dividers appeared quite early11 and were consistently used in most official inscriptions in Phoenicia12, Moab13, Zincirli14, Tell Fakhariyeh15, Tel Dan16, and Judah17 (The Royal Steward Inscription18, the Siloam Tunnel Inscription19). The best-known early examples of scriptio continua are the Gezer calendar (with one exception)20 and the Kuntillet ʕAjrud stone basin21. As rightly surmised by Naveh22, the lack of word-dividers in early NWS epigraphs is a feature of short and informal inscriptions, which may well be applied to the piece published presently.
11. Cross 2003, 213–215, fig. 32.1 and 32.2 (Qubur el-Walayda), Maeir et al. 2008, 53 (Tell eṣ-Ṣafi).

12. E.g. Naveh 1982, 52, fig. 43, 44; Rollston 2014a, 76–83, fig. 2–8. For relatively rare exceptions, usually from outside the Phoenician mainland, see Millard 1970, 5–6.

13. Naveh 1982, 64–65, fig. 55, 56; Dearman 1989, 307, fig. 1.

14. Naveh 1982, 5, fig. 45; Pardee 2009, 53, fig. 2; Lemaire, Sass 2013, 60, fig. 2.

15. Abou-Assaf et al. 1982, table (unnumbered); Rollston 2014a, 90, fig. 12.

16. Biran, Naveh 1995, 10, fig. 9.

17. See Millard 1970 and Naveh 1973.

18. Avigad 1953, pl. 9B, Naveh 1982, 68, fig. 59.

19. Naveh 1982, 68, fig. 60.

20. Naveh 1982, 63, fig. 54.

21. Meshel 2012, 76, fig. 5.3

22. Naveh 1973, 208.
20 A peculiar feature of the inscription is the ligature-like connection of the lower elements of two adjacent letters, occurring three times. This is a rare feature in NWS epigraphs, but some conspicuous examples are attested, primarily in the Siloam inscription23, but also in Kuntillet ʕAjrud24 and on some Hebrew seals25.
23. Naveh 1982, 68, fig. 60 (nun + qop, he + zayin, he + yod, bet + resh, kap + waw, bet + resh, mem + waw, mem + aleph, and especially bet + he, typologically quite similar to what we find in our inscription).

24. Meshel 2012, 88, 92, 95, 99, 101, fig. 5.26, 5.35, 5.37, 5.41, 5.43.

25. Naveh 1982, 69, fig. 61b (including two non-adjacent letters connected over a third letter in between).
21 In several cases, the letters are not positioned on the same line, but slanted or otherwise displaced, in an attempt to harmonize the orientation of certain signs to the neighboring ones. Thus, in the group of the first three letters only the dalet is located on the bottom of the inscription, whereas the zayin and the taw do not touch the bottom line, but are raised up to match the dalet’s triangle (fig. 7). This tendency, in all probability reflecting the engraver’s aesthetic ambition, is somewhat reminiscent of the horror vacui principle in Islamic art, which often influenced the design of Arabic inscriptions. Together with a marked leftward elongation of the “tails” of some letters, this feature gives the inscription a distinct ornamental pitch26.
26. A rather good comparandum is the middle part of the stone basin from Kuntillet ʕAjrud.
22 Most of the repeated letters are quite homogeneous in their shape (dalet, he, yod, kap). The most prominent exception is mem, the two specimens of which are rather dissimilar. To a smaller degree, the same is applicable to the three instantiations of nun.
23 3.2. Individual letters
24 Bet
25 No. 19. This is the angular type of bet with a circular head. The very elongated foot extends far to the left and seems to cross27 what looks like a remain of one more letter which we are unable to retrieve – most probably taw, its horizontal stroke and faint traces of the vertical shaft can be seen.
27. For the tail of bet crossing a neighboring letter see Vanderhooft 2014, 113.
26 No. 24. The head is shaped almost as a triangle with a straight bottom side and two other sides forming an arc; a long, almost horizontal foot, slightly damaged in the end, extends far to the left, almost touching the leg of no. 25. A very close parallel is found on the Kuntillet ʕAjrud stone basin28.
28. Meshel 2012, 76, fig. 5.2.
27 Dalet
28 No. 2, 23. Shaped as a triangle, the bottom is almost parallel to the baseline. Similar archaic dalets are attested, inter alia, on arrowheads29, Old Byblian royal inscriptions (PPG, Taf. I), the Manaḥat scherd30, the Gezer calendar31, the Tel Zayit abecedary32, the stele of Mesha33, the Kuntillet ʕAjrud stone basin (Renz’s type c, and more generally, group a–d). By the end of the 9th – beginning of the 8th century, the right side of the triangle began to be lengthened, forming a small leg34. However, the triangle-shaped dalet seems to have sporadically survived in the Hebrew script until the last quarter of the 8th century35.
29. McCarter 2008, 50, fig. 2, 2–4.

30. Hamilton 2014, 43, fig. 10b.

31. Naveh 1982, 63, fig. 54.

32. Tappy et al. 2006, 27, fig. 16, 17.

33. Naveh 1982, 64, fig. 55; Dearman 1989, 307, fig. 1.

34. Thus, this is the only form of dalet used on seals from the reigns of Uzziah (769–733) and Ahaz (733–727) (Avigad, Sass 1997, 57–58, no. 3B, 4, 5, 29). In the Phoenician and Aramaic epigraphy, similar developments take place from the 9th century (Schmitz 2014, 155; Rollston 2014b, 211; Balakhvantsev 2016, 20).

35. E.g. Renz 1995b, Taf. 14 (Jer (8)6).
29 He
30 No. 4. The vertical stroke of the he is slanted to the left36, three parallel horizontal strokes of the same length are adjoined to the vertical stroke, whose lower part is moderately long (Renz’s type o, and, more broadly, groups F, G and H). The slanting of the vertical stroke in the Hebrew monumental script is thought to appear in the 8th century37, but similar forms of he are, in fact, attested already on the stele of Mesha38 and the Amman Citadel inscription39. A more or less substantially left-slanted he further characterizes most of the Zincirli inscriptions40.
36. For the clockwise and counterclockwise slanting of the vertical strokes of the letters, we have adopted the perspective from the top point of the vertical. Therefore “slanted (or tilted) to the left” means that the vertical stroke extends from northwest to southeast.

37. Vanderhooft 2014, 114.

38. Naveh 1982, 64, fig. 55; Dearman 1989, 307, fig. 1.

39. Discussed in Cross 2003, 95–96.

40. Tropper 1993, 339, Abb. 3.
31 No. 13. The only difference from no. 4 is that the lower part of the vertical stroke is crossed by the foot of no. 12.
32 No. 20. Paleographic analysis is hampered by the fold of the paper. The letter is strongly tilted to the left, the vertical stroke does not seem to be prolonged beyond the bottom horizontal stroke41.
41. Similar shapes of he with a shortened vertical stroke (Renz’s type ff) are found on Hebrew bullae (Avigad, Sass 1997, 188, 200, 205, no. 457, 501, 524).
33 Waw
34 No. 21. A long diagonal shaft descending from right to left, slightly curved in its lower part; a short stroke extends to the left from ca. the upper third of the shaft. This У-shaped waw is atypical for Hebrew inscriptions42, where the Y-shape seems to predominate among the – admittedly few – early specimens (Renz’s type c(a))43. Elsewhere, comparable early У-shapes (and Ч-shapes, in all probability related) are found in the Gezer calendar (2:7, 5:7)44, the Shipiṭbaʕal inscription45 and the Amman Citadel inscription46. A fine archaic precedent is found on the Tell eṣ-Ṣafi bowl fragment47 and, as rightly observed by its editors, this was the letter shape adopted by early Greek alphabets48.
42. The closest approximation is apparently found in Renz’s Lak(8):17 (“korrigiert nach Photo”), see Renz 1995b, Taf. 6. To some extent comparable shapes can also be found on the ostraca from Arad no. 89–91, 95 (ibid., Taf. 12, the last quarter of the 8th century)

43. Vanderhooft 2014, 115.

44. Naveh 1982, 63, fig. 54.

45. Rollston 2008, 61, fig. 4; Sass 2017, 120–121.

46. See Cross 2003, 96.

47. Maeir et al. 2008, 49, fig. 8–9 and the discussion ibid. 52–53. In Eshel et al. 2002, chart 9, the two signs are listed as samples of yod, but on the preceding page (n. 1) the authors apparently stick to their earlier interpretation.

48. See Naveh 1982, 180–182.
35 Zayin
36 No. 1. An uppercase “H” rotated almost 90 degrees. The top and bottom lines are parallel, rather long and very close to each other (as opposed to the early Phoenician zayin with a very long central bar), the rest of the vertical stroke can apparently still be seen as a dot between the two lines49. This form of zayin (Renz’s types a and c) is documented by the arrowheads of the 11th century50, the Gezer Calendar51, the Tel Zayit abecedary52 and the stele of Mesha53.
49. A zayin with no vertical connecting stroke features on two Hebrew seals, see Avigad, Sass 1997, 50, no. 3 (the time of the king Uzziah, 769–733 BC) and 185, no. 444 (the time of the prophet Jeremiah).

50. McCarter 2008, 52 fig. 2, 2–4.

51. Naveh 1982, 63, fig. 54.

52. Tappy et al. 2006, 27, fig. 16, 17.

53. Naveh 1982, 64, fig. 55; Dearman 1989, 307, fig. 1.
37 No. 18. An uppercase “H”, slightly inclined to the right. The left vertical stroke is very massive, the right one rather thin. To our knowledge, there are no other instances of a “vertical” zayin in the NWS epigraphy54. Similar transformations are known for other letters, however: ḥet is rotated 90 degrees in the Gezer Calendar (5:3)55 and on arrowheads56, whereas lamed is mirrored on one of the arrowheads57 and rotated 180 degrees in Tell Fakhariyeh58. Last but not least, four out of six specimens of gimel on Shipiṭbaʕal inscription (KAI 7) feature their shorter stroke markedly turned to the right instead of the usual leftward orientation59.
54. One will not lose sight of the fact that “H” is the normal shape of the “South Semitic” zayin (cf. Cross 2003, 223).

55. Cf. Maeir et al. 2008, 50.

56. Cross 2003, 207.

57. Deutsch, Heltzer 1999, 9, fig. 121.

58. Abou-Assaf et al. 1982, table (unnumbered); Rollston 2014a, 90, fig. 12.

59. Cf. McCarter 1975, 35–36.
38 Ḥet
39 No. 17. Two verticals, slightly slanted to the right and connected by three short horizontals. The verticals do not cross the two lower horizontals, the lower angles of the “box” are, moreover, slightly rounded. This archaic box-shaped ḥet is attested on the ʕIzbet Ṣarṭah sherd60, arrowheads from the 11th century61, the ʕAzarbaʕal inscription62, the Kefar Veradim bowl63, the Nora fragment64, the Manaḥat jar fragment65, the Ḥorvat Rosh Zayit jar fragment66, the Gezer calendar and the Ras ez-Zetun ostracon67. Renz’s putative later attestations68 are generally unreliable69, but a closer approximation may perhaps be seen on the Ammonite seal no. 110270.
60. Rollston 2014a, 74, fig. 1, with Cross’ discussion in 2003, 222.

61. McCarter 2008, 52, fig. 2, 3–4.

62. Rollston 2014a, 76, fig. 2.

63. Rollston 2014a, 77, fig. 3.

64. Naveh 1982, 40, fig. 35; Cross 2003, 261, fig. 37.2.

65. Stager 1969, 51, fig. 3; Hamilton 2014, 43, fig. 10b.

66. Hamilton 2014, 45, fig. 11e.

67. Renz 1995b, Taf. I, no. 2.

68. Renz 1995c, 141–142.

69. Cf. his own remark in Renz 1995a, 37 (“Ḥ als Vierstrichform, in einer Form ohne Überstände, die einerseits im Grunde nur im 10. Jhdt. belegt ist, anderseits auch noch nicht eindeutig hebr. ist”).

70. Herr 2014, 177, fig. 1.
40 Yod
41 No. 7. A vertical stroke inclined ca. 45 degrees to the left, from which two parallel, rather prominently oblique lines extend to the left at the top and in the center. The lower one is slightly longer than the upper. Another short parallel stroke, also oblique, starts from the bottom point of the vertical stroke to the right. The vertical stroke of yod rotated counterclockwise is atypical for Hebrew epigraphs where the shaft of yod is usually vertical or slanted to the right71. Rare parallels are found in Khirbel el-Qom inscriptions72 and on lmlk jar handles73. The choice of this unusual form could be accounted for by aesthetic concerns of the engraver, who strived to harmonize the stance of the adjacent letters. At the same time, yods with the vertical stroke slanted to the left are common in Phoenician epigraphy. The earliest examples are provided by the Kilamuwa inscription (PPG, Taf. I, 10; ca. 825), the gold medallion from the Douïmès necropolis in Carthage (8th or 9th c.)74 and the inscription from Hasan Beyli (PPG, Taf. I, 11; ca. 715). Later on, this form of yod becomes predominant in the Phoenician script (PPG, Taf. I, 13–14; Taf. II, 4–14; Taf. III, 4–14).
71. Vanderhooft 2014, 116.

72. Renz 1995b, Taf. 15.

73. Renz 1995b, Taf. 18.

74. The 8th century date is adopted in PPG (Taf. III, 3). For the 9th century date see Pilkington 2019, 6. A thorough paleographic reexamination of the inscription led P. Schmitz to date it to 800–775 (Schmitz 2008, 171)
42 No. 22. The letter is peculiar insofar as the lower half of the shaft does not straightly continue the upper part, but is heavily slanted leftwards and almost coincides with the second horizontal (being, in fact, almost invisible).
43 No. 31. The bottom horizontal stroke is connected to no. 30 (kap), forming a curvative ligature.
44 No. 33, 36. Connected to the previous letters more or less in the same way as no. 32 is connected to no. 31.
45 Kap
46 No. 26. A long downstroke, slightly curved down to the left. Two side-branches stem symmetrically from a single point in the upper part of the shaft (Renz’s type d). This type of kap, reliably attested on Hebrew epigraphs75, is the standard one on the stele of Mesha and the Kerak inscription76. A strikingly similar early parallel comes from the Tel Zayit abecedary77, and cf. already the kaps on the ʕIzbet Ṣarṭah ostracon78.
75. Close early parallels come, e.g., from Kuntillet ʕAjrud (Renz 1995b, Taf. 3).

76. Vanderhooft 2014, 117, see further ibid. 108, fig. 1; Naveh 1982, 64–65, fig. 55–56.

77. McCarter 2008, 52, fig. 3.

78. Naveh 1982, 37, fig. 31 and Hamilton 2014, 48, fig. 13.
47 No. 30. This kap forms a quasi-ligature with the adjacent yod (no. 31): the leg continues without interruption into the bottom right stroke of yod.
48 No. 35. Like no. 31, connected to the adjacent yod (no. 36).
49 Lamed
50 No. 28. A curved long vertical stroke with a widely open loop at its bottom.
51 No. 15. This unusual lamed features a closed loop in its lower part (cf. Renz’s type f). Several examples of lamed with a near to closed loop are found in Kuntillet ʕAjrud79, with some further comparable attestations on Samaria ostraca80. It is not without interest that lameds with a heavily curled, (nearly) closed loop are very common in “Early Alphabetic B” epigraphs81.
79. Renz 1995b, Taf. 3 (Pithos 1), Taf. 4 (Eingravierte Worte), cf. “the lamed with a rounded base” (Meshel 2012, 81).

80. Renz 1995b, Taf. 6, 10.

81. Such as the Tell el-ʕAjjul cup (Hamilton 2014, 35, fig. 5), the Tell eṣ-Ṣafi bowl fragment (ibid., 39, fig. 8), the Qubur el-Walaydah sherd (Cross 2003, 214, fig. 32.1), the Lachish ostracon (ibid., 94, fig. 46.2), the 2001 Beth Shemesh inscription (McCarter et al. 2011, 188, fig. 5, and 190), the 2014 Lachish jar sherd (Sass et al. 2015, 235, fig. 2, and 242). See further Cross 2003, 214–215; Maeir et al. 2008, 51 and 53.
52 Mem
53 No. 6. A straight diagonal shaft, slightly damaged in the upper part, stretches down from right to left. A rather shallow W-shaped head adjoins to its upper end, stretching to the left. This is, essentially, Renz’s type e, for which he, however, provides no examples. The leg exhibits no curvature (let alone curling), characteristic of “standard” Moabite and Hebrew mems (Renz’s type f).
54 No. 14. The head is, peculiarly, ω-shaped and, thus, quite different from no. 6. Comparable forms are attested on Hebrew seals from the 8th century82 and on the ostracon from Tell el-Qaḍi83 (cf. Renz’s type k).
82. Avigad, Sass 1997, 159, no. 377; Deutsch, Heltzer 1999, 34–35, fig. 130; Deutsch, Lemaire 2000, 71, no. 65.

83. Renz 1995b, Taf. 9.
55 Nun
56 No. 16. A short upper vertical stroke slightly tilted to the right, a horizontal bar, and a curved downstroke stretching to the left and forming a well-pronounced small “foot”. No exact precedents are at hand. The closest parallel in Renz’s typology is type c, but some features make it closer to his type k.
57 No. 34. Typologically rather dissimilar from no. 16. The upper vertical stroke is longer and its juncture with the horizontal bar is much more rounded. The latter, in turn, is not straightly horizontal but concave, protruding downwards. The lower vertical stroke is curved, not angular, and features a sharp end instead of the “foot”. It is hard to say whether a sort of “horn” above the angle formed by the horizontal bar and the lower vertical belongs to the letter or is due to a visual effect of uneven paper. If the former is correct, the letter would come close to the nun of the Moabite Kerak inscription84 (Renz’s type f)85. A more angular, but structurally rather exact early precedent is found on Tel Rehov ostracon no. 386.
84. Naveh 1982, 65, fig. 56.

85. Within this approach, the letter could, theoretically, be also interpreted as an Ammonite-style bet (or reš) with a very broad “open head” (Herr 2014, 177, fig. 1). Since no similar development is seen anywhere else in the inscription, this is a rather unlikely possibility.

86. Mazar 2003, 178, fig. 5.
58 No. 38. The three upper strokes seem to form a good right-angled zigzag. A small “horn” above the juncture of the horizontal bar and the lower vertical (as in no. 34) is not very well seen, but cannot be ruled out. As clearly seen on fig. 8, the second vertical (crossed by the fold of the squeeze) displays a bottom curvature ending with a small “foot”. A similar shape of nun is found on the 9th century Arad ostracon no. 76 (l. 3)87.
87. Renz 1995b, Taf. II, no. 1.
59 ʕAyin
60 No. 27. A regular circle without interruptions and no “pupil” in the middle.
61 Pe
62 No. 10. The form of pe corresponds to Renz’s types h and i (a relatively long, somewhat curved extension in the bottom)88. The upper left extension merges with the right upper stroke of the following letter. Typologically, this shape is comparatively late89, but a clear-cut early precedent is found on the 9th century Pithos B from Kuntillet ʕAjrud90.
88. Cf. the examples in Renz 1995b, Taf. 18, 20 and Vanderhooft 2014, 108.

89. Rollston 2014b, 223–224.

90. Renz 1995b, Taf. IV, no. 1.
63 Šin
64 No. 5. Four strokes are shaped in the form of a “W”. All strokes have approximately the same length, like šin in the Gezer calendar and Moabite stelae (Renz’s types a, b, d)91.
91. Vanderhooft 2014, 108, fig. 1, 1–3.
65 Taw
66 No. 3. A cross with a long downstroke slanted down to the left and a short horizontal stroke neatly parallel to the base of the inscription. The left extremity, somewhat shorter than the right one, seems to feature a thin raised end. The very short upper element of the downstroke looks a bit rounded and somewhat detached from the horizontal bar, likely due to a damage on the stone92. This form of taw, not characteristic of Hebrew inscriptions93, is nevertheless sporadically attested in this corpus, notably in the Khirbet el-Qom cave inscription94 and elsewhere95. Similar forms of taw96 are best attested in Zincirli97 and Sefire98, to some extent also in Tel Dan99. For this form of taw in the Phoenician script see KAI 30 (PPG, Taf. I, 8; Cyprus, 9th c.), Kilamuwa (KAI 24; PPG, Taf. I, 10; ca. 825), and Karatepe (PPG, Taf. I, 12; KAI 26, ca. 700). The extreme slanting of the downstroke may have originated out of the general tendency of the engraver to accommodate adjacent letters to each other.
92. The shape of the upper part of the letter is curiously reminiscent of the “Sun-child” image (as on the Revadim seal, Cross 2003, 300, fig. 48.1).

93. Vanderhooft 2014, 120.

94. Vanderhooft 2014, 108, fig. 1, 11.

95. Renz 1995b, Tf. 2.

96. McCarter 2008, 56.

97. Naveh 1982, 55, fig. 45.

98. Dupont-Sommer, Starcky 1958, pl. XIII.

99. Athas 2003, 133, table 4.19.
67 3.3. Problematic segments
68 No. 8
69 This segment looks exactly like a classical Phoenician/Old Aramaic ṣade, but mirror-faced: a long diagonal shaft stretching down from right to left, to which, by its upper third, a three-stroke zigzag is joined. At present, no clear identification can be proposed. The following options are, theoretically, under consideration:
70 1. A mirror-faced ṣade as described above;
71 2. An unusual mem whose fourth (rightmost) stroke has merged with the shaft (somewhat similar to Renz’s type e, for which he gives no examples);
72 3. Since the head is shaped very much like šin (no. 5), one could surmise that the letter is actually a šin, whereas the rest of the vertical stroke belongs to the following segment (itself paleographically problematic);
73 4. The shaft plus the rightmost stroke stemming from it are similar to the У-shaped waw (no. 21). This would allow one to attribute the remaining two short strokes to the following segment, then perhaps identifiable as ṣade (see below).
74 No. 9
75 Paleographic analysis of this (and adjacent) segment(s) (fig. 9) is hampered by the fold of the squeeze. What can be seen looks like a straight diagonal shaft stretching down from left to right, a longer stroke stemming upwards to the right from its top (or close to it) and possibly crossing the shaft, and a parallel, very short, but clearly visible stroke stemming from its middle.
76 Taken at face value, this shape is best compatible with the later Hebrew aleph, represented by Renz’s types u, v, w, x (note especially the slanted shaft in w and x). This shape is attested since the second half of the eighth century100, notably, on the Arad ostracon no. 51101 and the Siloam tunnel inscription102. It is not without interest to observe, incidentally, that two alephs consisting of a vertical shaft, slightly tilted to the right, and two almost parallel horizontal strokes extending from it to the right appear to be attested on the Byblos “clay object” B, dated by Cross to mid-eleventh century103.
100. Aharoni 1981, 131.

101. Renz 1995b, Taf. 11.

102. Renz 1995b, Taf. 13.

103. Cross 2003, 336, fig. 53.6. Interestingly, the horizontal strokes only slightly cross the vertical in the second specimen and do not cross it at all in the first one. See further Cross, McCarter 1973, 8.
77 Similar looking forms of waw come from the Arad ostraca (e.g. no. 73) of the 8th century104 and, mutatis mutandis, other Hebrew epigraphs (Renz’s “Dreistrichform”, types h, i, j). Since a radically different У-shaped waw is reliably attested in the inscription (no. 22), this identification is not very promising.
104. Renz 1995b, Taf. 7.
78 As pointed out above in connection with no. 8, if the latter is interpreted as a waw and the remaining part of its head is joined to the present segment, this would look like a well-shaped ṣade – the closest parallel is found in the Amman Citadel inscription, dated 9 c. BC105.
105. Renz 1995b, Taf. 2.
79 No. 11
80 A long downstroke, slightly curved down to the left and crowned by three symmetrical short strokes. The right short stroke is merged with the upper element of the preceding pe. The shape comes very close to what has been identified above as kap. However, seen from some angles at least, the middle stroke of the crown appears to be continued down by a straight line touching the upper end of the “foot” of the preceding pe, thus creating a taw-like image (†), somehow superimposed on a kap.
81 No. 12, 25 and 32
82 As far as we can see, a proper distinction between bet and reš is hard to achieve in each of the three cases.
83 No. 25 features a very elongated leg stretching far to the left and almost parallel to the base of the inscription. In principle, this is scarcely compatible with reš, whose shaft is expected to be straightly vertical (or, rather, even slanting to the right) and have no “tail”. At the same time, the letter is quite dissimilar from the preceding clear-cut bet, whereas two repeated bets are unlikely to occur here on linguistic grounds. It stands to reason, therefore, that we are dealing with an unusual reš whose extra-long leg is to be accounted for by the aesthetic preferences of the engraver. Still another possibility is to identify this segment as a qop: the head is practically circular and the shaft appears to be joined to it centrally rather than from the right (although a loss of stone under the head to the right is a feasible possibility). The unusual elongation of the shaft would remain to be explained, anyway.
84 No. 12 features a circular head with a downstroke stemming from its right side down to the left. The extremity of the downstroke is slightly bent to the left and intersects the vertical stroke of the next letter (he). A reš is perhaps more probable than a bet, which, however, can hardly be excluded.
85 No. 32 forms a ligature with the following he so that the nature of its vertical stroke cannot be evaluated objectively. A reš and a bet appear to be equally possible.
86 No. 29
87 The vertical shaft is moderately slanted to the right. The upper horizontal stroke extends to both sides of the shaft, the right segment is much longer than the left one. The vertical shaft does not go beyond the upper horizontal. The lower horizontal stroke, stemming from the vertical but not going beyond it to the left, is parallel to the upper bar and may be slightly longer than it. Taken at face value, this shape could be identified with a mature Hebrew aleph of the Siloam tunnel inscription (cf. no. 9 above), in which, however, the vertical shaft does clearly go beyond the upper bar106.
106. On this important feature of the “classical” Hebrew aleph see Rollston 2003, 160.
88 Some photos give the impression that the letter has a third (middle) horizontal bar, less prominent than the other two and sometimes looking like an in-between protrusion of the paper. Upon this reading, the letter is be interpreted as a samek, in which the middle and bottom horizontal strokes had lost their left continuations before the squeeze was made. Indeed, a brighter oval spot can be detected exactly where the two small left strokes would be expected to be located, perhaps reflecting a chopped-off surface.
89 No. 37
90 The letter looks like a kap with four upper strokes instead of the regular three. No convincing interpretation is at hand. A certain similarity to the Egyptian hieratic numbers “400”107 or “600”108 is to be noted. The use of the Egyptian hieratic numbers is known from the Hebrew inscriptions on ostraca, weights and bullae in the preexilic period109.
107. Möller 1909, No. 635; Wimmer 2008, 231.

108. Möller 1909, No. 637; Wimmer 2008, 233.

109. Avigad, Sass 1997, 177; Deutsch, Heltzer 1999, 64, 67.
91 3.4. Paleographic evidence for a chronological and geographical setting
92 The chronological and geographical perspective of our paleographic analysis is of necessity rather uncertain. This is due, first of all, to the well-known fact that early NWS epigraphs (let alone, lapidary inscriptions) from areas other than Phoenicia are poorly represented110. Hot debates on the chronological, geographical and, for that matter, linguistic nature of such documents as the Gezer calendar or the Tel Zayit abecedary, in which leading authorities in the field have often come to virtually opposite conclusions, are telling witnesses to this effect. It is upon this difficult background that the following, very tentative observations are to be evaluated.
110. “Surviving specimens of the tenth-century inland script are relatively rare” (Tappy et al. 2006, 28); “There are simply not enough inscriptions on which to base conclusions” (Schniedewind 2005, 405); “A gap of a century or a century and a half filled only by one substantial inscription” (Cross 2003, 340).
93 3.4.1. Chronological attribution
94 The inscription features some indisputably archaic (or archaizing) features:
95 1. A clear-cut box-shaped ḥet, hardly ever attested later than the 10th century;
96 2. Two legless dalets111, “an important diagnostic letter”112.
111. Tappy et al. 2006, 33.

112. Vanderhooft 2014, 114.
97 Less certain, but still probable archaic elements are a rotated zayin, a lamed with a closed loop, perhaps a mirrored ṣade.
98 These are opposed to a few letters whose shape is, by the accepted standards, typologically more advanced:
99 1. Two mems with a horizontal zigzag, as opposed to the archaic Phoenician mem with a vertical zigzag. Note, at the same time, that in each of the two specimens the vertical shaft is rather short, straight and uncurved – a feature usually considered archaic113;
113. Tappy et al. 2006, 37.
100 2. None of the three examples of nun can be said to correspond to the classical perception of the archaic nun – an angular zigzag of three strokes more or less equal in length;
101 3. The inscription does not feature the bowl-shaped head of waw, usually considered archaic114. A У-shaped form is used instead;
114. Tappy et al. 2006, 33; Rollston 2014a, 87.
102 4. The pe with a long, curved leg ending with a well-pronounced “foot” is typologically innovative.
103 Can the two trends be harmonized? Are they expected or likely to be reflected in one monument? In our view, the answer to these questions may well be positive.
104 Indeed, according to a growing consensus, certain paleographic shapes traditionally considered to be characteristically “old”/“archaic” vs. “recent”/“innovative” can, on the one hand, coexist as variants from the earliest periods onwards115 and, on the other hand, be simultaneously attested in a single epigraph, possibly (but not necessarily) due to the archaizing efforts of the scribe116.
115. “In the analysis of the typological developments of these scripts, the simultaneous, overlapping occurrence of various letter types must be taken into account ... Different letter types were in use at the same time in different parts of Canaan” (Maeir et al. 2008, 61).

116. “Preservations of typologically older forms are to be anticipated ... at times” (Rollston 2014a, 87); “In some cases, letter variations can be seen even in the same inscription” (Maeir et al. 2008, 61); “With its four-barred ḥet and relatively advanced mem and nun, [this inscription] is difficult to associate comfortably with either the mainstream Phoenician tradition or with its inland, southern development” (Tappy et al. 2006, 29 on the Tel Zayit abecedary); “It is hard to say whether this difference, and a few more ... have chronological significance or represent discrete, contemporary scribal traditions” (Finkelstein et al. 2008, 6), “an artificial script – no inner order” (Sass 2017, 131).
105 Thus, a fully-fledged “classical” ḥet with vertical strokes crossing the upper and lower horizontals from the left and right sides respectively is attested as early as in the Tel Zayit abecedary117, the Tel Batash ostracon118, the Ophel pithos119, the Byblos “clay objects” A and B120, a jar from Tel Reḥov121. To these one can add the shapes with two verticals crossing the horizontal on the Bet Shemesh gaming board122 and two verticals each crossing the upper and lower horizontals from the right on the Raddana handle123.
117. Tappy et al. 2006, 33–34.

118. Hamilton 2014, 45, fig. 11c.

119. Mazar et al. 2013, 41, fig. 3.

120. Cross 2003, 336, fig. 53.6.

121. Ahituv, Mazar 2014, 47, fig. 8.

122. Sebbane 2016, 640, fig. 19.1.

123. Cross 2003, 331, fig. 53.1. For the coexistence of these shapes with the box-shaped ḥet see ibid., 222, 341; McCarter 2016, 647.
106 The innovative mem with a more or less horizontal head is attested as early as the Tel Zayit abecedary124 and has been qualified as having “a surprisingly advanced appearance ... in the context of the rest of the abecedary, with its numerous archaic forms”125. Two mems with clear-cut horizontal heads and, moreover, very long legs markedly curved to the left at their bottoms are seen on the Tel Rehov jar fragment (no. 2), dated by the editor to the 9th century126. These are very similar to later “classical” Hebrew types. In the same article, Mazar published another jar fragment, with the same chronological attribution, where the mem even features a small “foot”127. At the same time, an archaizing mem with a vertical zigzag is found as late as on the Royal Steward inscription128.
124. Tappy et al. 2006, fig. 16, 17 and cf. McCarter 2008, 54.

125. Tappy et al. 2006, 36–37. Cf. also Maeir, Eshel 2014, 210, fig. 8 (a long-tailed mem with a horizontal lead from a Tell eṣ-Ṣafi jar inscription).

126. Mazar 2003, 175, fig. 3, 177, fig. 4, with discussion.

127. Mazar 2003, 178, fig. 3, 179, fig. 4.

128. Vanderhooft 2014, 117.
107 A very clearly shaped “foot” of the nun is found on the ʕIzbet Ṣarṭah ostracon129.
129. “As Kochavi has remarked, the curved lower leg is unparalleled and, I should add, probably misdrawn” (Cross 2003, 223). Since the letter in this shape appears thrice on the ostracon, it is hard to see how it could be “misdrawn”.
108 The У-shaped waw of the inscription is neither the early Phoenician bowl-headed type, nor the Y-type characteristic of some early Hebrew epigraphs. Still, it is hardly possible to say that the У-shape (and the likely related Ч-shape) are recent: as shown above, they are attested as early as the Gezer Calendar and the Tell eṣ-Ṣafi bowl fragment130.
130. Coexistence of the two major types of taw (+ and × shaped) in early NWS epigraphs is discussed in Maeir 2008, 52.
109 The ʕAzarbaʕal spatula features six šins and nearly all of them belong to different, sometimes markedly dissimilar types, whereas several types of waw and ḥet clearly coexist in the Gezer calendar. Similar examples can easily be multiplied.
110 In any case, the inscription does not display any of the diagnostic features of the “classical” Hebrew epigraphic style131. Only the hypothetic aleph (no. 29) could point in this direction132, but this identification is far from obvious.
131. Like those gathered in Schniedewind 2005, 407.

132. Vanderhooft 2014, 112.
111 One may summarize that hardly any decipherable sign of the inscription should be later than the ninth century. Since the ḥet (probably also the dalet) are typologically older than that, a certain degree of paleographic archaization is likely to be surmised. And since, in our view, archaization is best not to involve too long timespans, a mid-9th century seems a paleographically justified approximate dating.
112 3.4.2. Geographical setting
113 As far as we can see, a few letter shapes point to a Southern Canaanite (rather than Phoenician) setting of the inscription:
114 1. The kap is decidedly non-Phoenician: to the best of our knowledge, a combination of a symmetrical trident shape with a long, curved leg is never attested in this tradition133. Conversely, this is a deeply rooted Palestinian form reflected in ʕIzbet Ṣarṭah, Tel Zayit, the Moabite stelas and in early Hebrew134;
133. McCarter 2008, 54.

134. Demsky 1986, 191; Tappy et al. 2006, 35–36; Cross 2003, 222–223; Hamilton 2014, 47–49. Contrast Rollston 2008, 81–89 and cf. a summary in Vanderhooft 2014, 109.
115 2. Each of the three nuns demonstrate, in a more or less pronounced way, “the tendency characteristic of the later Hebrew nun ... for the lengthening shaft to bend from right to left and then to curl upward at the end”135. This feature, characterizing this letter in most of the 9th century Hebrew (and Moabite) epigraphs, is practically unattested in Phoenician;
135. Tappy et al. 2006, 37.
116 3. The inscription features a characteristically Hebrew pe, with a long, well pronounced “foot”, very similar to the shape first attested in the Kuntillet ʕAjrud abecedary;
117 4. The broad zayin with two long horizontal bars and a very small space between them (letter no. 1, if correctly identified) contrasts sharply with the early Phoenician I-shaped zayin and comes close to the shapes attested in the Gezer calendar and the Tel Zayit abecedary136.
136. McCarter 2008, 53–54.
118 Since there is hardly any shape that would radically contradict a Southern attribution of the epigraph, we may cautiously conclude that the cumulative evidence seems to locate the origin of the inscription in Southern Canaan (Palestine) rather than in Phoenicia.
119 4. Towards a linguistic and philological analysis
120 For obvious reasons, the interpretation of the text is faced with severe difficulties. In the absence of word-dividers, the narrow inventory of NWS graphemes (some of them, moreover, paleographically uncertain) can be combined into many alternative strings which, while more or less meaningful per se, may turn out quite meaningless in context. Besides, the text is, in all probability, typologically unique and does not seem to feature any proper names such as anthroponyms, theonyms or geographic designations137.
137. If no. 8 is read as W, the ensuing sequence ŠMYW could represent a Yahwistic personal name *šǝmūyāw. Apart from its uncertain paleographic perspective, this solution does not seem to yield any promising avenue of interpretation for the text as a whole.
121 With these limitations in mind, most specialists would probably agree that any attempt at a definitive understanding of the inscription is deemed to be a high-level linguistic and philological challenge. Our considerations below are, thus, of necessity preliminary and primarily intended to open the debate rather than to offer safe conclusions.
122 By far the most reliable string of letters that can be individualized is the sequence ʕL (no. 27–28). Both letters are paleographically straightforward, and since neither KʕL, nor ʕLʔ, nor ʕLS138 are likely to be meaningful in NWS, K and ʔ/S should be separated from ʕL as belonging to what precedes and what follows respectively.
138. To be sure, a form of ʕls ‘to rejoice’ (HALOT 836) is theoretically possible, but quite unlikely if only because of the very late attestations of this root in the Old Testament.
123 The ensuing sequence ʕL could be equated to the Hebrew preposition ʕal or to a form derived from the verbal root ʕly ‘to be high, to go up’ – a perfect or an imperative. The first and third options would agree with the “standard” orthography of epigraphic Hebrew139; the second would contradict it, but it is, of course, hard to say to what extent the “classical” h-writing was standard at such an early date as the one to be postulated for our text140.
139. No IIIh imperatives are attested in the Hebrew epigraphic corpus, but the h-less jussives yhy and w-yrʔ (Gogel 1988, 96; Renz 1995c, 217, 232) presuppose that the imperatives were also spelled without -h. The Moabite Mesha stele shows the same pattern (Dearman 1989, 104, 116). For a spelling without -h outside the short form of the prefix conjugation, cf. qn ʔrṣ ‘possessor/creator of the earth’ (Renz 1995b, 198).

140. Spellings without h are, of course, standard in (post-Old Byblian) Phoenician.
124 If no. 29 is read as aleph, the only feasible option is to join it to the next letter, individualizing the resulting sequence ʔK as the Hebrew adversative particle ʔak ‘yet, but’. This would exclude the prepositional reading of the preceding ʕL.
125 If no. 29 is read as samek, possible interpretations of SK are also few in view of the rarity of s in Hebrew.
126 By far the most appealing one is the imperative of *nsk ‘to pour out’, attested in Hebrew, Phoenician and Ugaritic in one of the two main semantic ramifications: ‘to make a libation’ and ‘to cast metals (for molten images)’ (BDB 652, DNWSI 735–736, 664–665, DUL 635–636). By Biblical standards, the imperative without n would not be normal here141, but Ugaritic does provide an impressive piece of evidence in favor of such an early NWS form exactly for the root under scrutiny: sk šlm l kbd ʔarṣ ‘Pour out peace into the bosom of the earth!’ (KTU 1.3 iii 16)142. On this reading, the preceding ʕL can only be interpreted as the imperative of ʕālā ‘to go/come up’.
141. Gesenius 1910, 173. But cf. such exceptions as gōšū and gōšī from nāgaš/yiggaš (ibid.).

142. Cf. Tropper 2012, 629 for a diachronic evaluation. Elsewhere in the Ugaritic corpus, nsk is used, inter alia, about rain poured down from the heaven: ṭl šmm tskh || rbb nskh kbkbm ‘Dew that the heaven pours down, || drizzle poured down by the stars’ (KTU 1.3 ii 40–41).
127 Within an alternative, perhaps less likely approach, we could deal with a substantive sk, preceded by the preposition ʕl and, perhaps, followed by -y, either the masculine plural construct ending or the 1 sg. pronominal suffix. The only promising candidate for this identification is Hebrew sōk, sukkā ‘hut, refuge’ (HALOT 753), whose etymological s (rather than a late reflex of *ś) seems to be assured by Ugaritic sk ‘den, cove’ (DUL 745).
128 Back to the linguistically unlikely sequence KʕL (no. 26–28). After ʕL is separated, the remaining K can either be joined to a preceding form or to be analyzed as an isolated word of its own.
129 What kind of “preceding form” could be posited here?
130 If no. 25 is read as qop and joined to the preceding (reliable) bet (no. 24), the ensuing BQ does not look promising, and even less so BQK. If it is read as reš (rather likely), an Aramaic-like BR(K) ‘(your) son’ could theoretically be posited – or, at least superficially more likely, a form of brk ‘to bless’. Both solutions leave the preceding, quite secure dalet (no. 23) without any promising parsing.
131 It stands to reason, therefore, that the dalet is to be integrated into the word under discussion, yielding a form of dbr ‘to speak’ or dbq ‘to stick’ – both well attested, reliable Hebrew roots. A natural step forward is to further integrate the preceding, rather unassailable WY (no. 21–22), thus producing a 3 sg. m. form of the prefix conjugation form with a waw consecutive, equivalent to Hebrew wa-yyədabber ‘he spoke’ or wa-yyidbaq ‘he stuck to’. These (especially the former) are, intuitively, quite suitable readings, which, alternative parsing options pending, are best accepted as a working hypothesis143.
143. In principle, Y (no. 22) can be joined to D (no. 23) to produce yād ‘hand’, but we were not successful to propose any further development for this line of thought.
132 Since K as a 2 sg. object suffix can scarcely belong to either wa-yyədabber ‘he spoke’ or wa-yyidbaq ‘he stuck to’ (none of the two verbs is likely to be used with a direct object), we are compelled to interpret it as a one-letter independent lexical segment. For this, three theoretical possibilities are at hand:
133 1. The conjunction kī, which would not fit its standard orthography in the Hebrew inscriptions and the Old Testament (ky), but would fully correspond to the Phoenician norm (k) (DNWSI 497);
134 2. The comparative preposition kǝ-, orthographically blameless, but hard to coordinate with the following ʕL whatever it can mean;
135 3. The deictic adverb kō ‘here’ or ‘thus’ (HALOT 461), so far unattested in Hebrew inscriptions (but probably expected to be written with a final -h).
136 For the time being, the third solution appears the most promising. On this reading, kō may belong to either what precedes or what follows.
137 If the first option is preferred, the suitable meaning is ‘thus’: ‘and he spoke thus’. This is an appealing solution because in the Old Testament kō is routinely used about speech events. Admittedly, the verb employed in such statements is ʔāmar rather than dibbär (BDB 462, meaning 1a), but this is, perhaps, not a crucial obstacle.
138 If the second option is accepted, kō would be used as a locative adverb modifying the following imperatives: ‘come up here’.
139 Let us now turn to the paleographically reliable sequence HMLNḤZBH (no. 13–20). Discarding the unlikely possibility that H (no. 20) actually belongs to the following WY (to yield a form of Aramaic hwy ‘to be’), we must join it to the preceding B. Since neither ZBH nor ḤZBH can yield anything of interest as a verbal or nominal form, we are compelled to interpret BH as a separate sequence, in all probability identical to Hebrew bō ‘in it’ in its – fully expected – early h-orthography.
140 The preceding Z (no. 18) – not unlike its possible antecedent (no. 1) – could be easily identified with a reflex of the Proto-Semitic deictic/relative element *ḏV. Combined with the following BH, that would yield an acceptable and even promising sequence: ‘which (is) in it’. In this case, the preceding Ḥ (no. 17) must belong together with N (no. 16), in all probability as a form of *nwḥ ‘to have rest’. Since the preceding L (no. 15) can hardly be joined to either M or HM (no. 13–14), it has to be joined to NḤ as the purpose preposition. The latter would then represent an infinitive construct (= Hebrew lā-nūaḥ). As for the remaining HM (no. 13–14), it must represent the 3 pl. m. personal pronoun (= Hebrew hēm), either independent or suffixal. While not immediately yielding an acceptable syntagm, this line of interpretation is not to be neglected in the future discussion of the text.
141 Within a different, perhaps preferable approach, ḤZ (no. 17–18) is identified with the Hebrew verbal form ḥāzā ‘he saw’. This is appealing since this verb is attested in the idiom ḥāzā bǝ- ‘to look on intensely’ (BDB 302, meaning 1c). This reading would produce out of no. 17–20 a syntactically and semantically promising sequence: ḤZ BH ‘he saw him/her/it’, ‘he looked at him/her/it’. The preceding HMLN could theoretically be equivalent to Hebrew ha-mmālōn ‘the lodging place’ (HALOT 588), which, however, does not immediately yield any sense here. It is, therefore, more promising to divide it into HM and LN, interpreting the former as a 3 pl. m. personal pronoun (see above) and identifying the latter with Hebrew lānū ‘for us’144.
144. Judging from the nḥnw ‘we’ (Gogel 1998, 154), a final waw would then be expected in the spelling, but it is hard to see to what extent this very late (6th century) example can be relevant for the present purpose. The Phoenician norm is, of course, without -w.
142 It is now time to turn to the initial part of the inscription (no. 1–12, whose analysis is hampered by several grave paleographic uncertainties (Z?DTHŠMY??PK?B?).
143 As pointed out above, zayin seems by far the most likely reading for the first letter of the inscription, even if yod cannot be entirely ruled out. As long as the reading Z is accepted145, the letter could again be identified with PS *ḏV ‘this’, a rather appropriate initial word for an inscription (‘This (is) ...’). This solution is, however, difficult insofar as the following DT (or DTH) cannot be identified with any known NWS nominal lexeme. It stands to reason, therefore, that Z is to be jointed to the following DT (no. 2–3), yielding ZDT. This is vividly reminiscent of zdh in line 3 of the Siloam tunnel inscription, where the word occurs in the phrase ky.hyt.zdh.bṣr, intended to explain why the tunnelers were able to hear each other’s voices at the final stage of the construction. The exact meaning of the word is hard to establish, ‘crack’ and ‘resonance’ being the two most widely discussed options146.
145. If the first letter is Y, the ensuing sequence YDT could be identified with ydt (= Hebrew yādōt) ‘monuments’ from the Jerusalem Ostracon 1 (Dobbs-Allsopp et al. 2005, 207–209; Renz 1995a, 310–311). We have no clue to the resulting contextual interpretation. The well-attested Phoenician word ʔdt ‘lady’ cannot be an option here since the first letter is not an aleph even in a most theoretical perspective.

146. The literature on zdh is borderless, cf., among others, KAI II 187; Renz 1995a, 184–185; DNWSI 306; Dobbs-Allsopp et al. 2005, 503.
144 If ZDT stands for a feminine substantive in the construct state (≈ Hebrew *zēdat), we have to look for its nomen rectum in the following letters, and the immediately following H, easily identifiable with the definite article, is quite favorable for such a possibility. The main obstacle is the paleographic identification of no. 8.
145 If no. 8 is read as mem, the sequence HŠMYM (no. 4–8) can hardly be anything but Hebrew ha-ššāmayim ‘the heaven’. Joined with ZDT, this would give ZDT HŠMYM = *zēdat ha-ššāmayim ‘crack (in) the heaven’ or ‘resonance (in) the heaven’. The former would provide a fine parallel to the famous ‘rift in the clouds’ in the Ugaritic Baʕlu Epic147. It would likely place the inscription in a mythological context and/or connect it with rainwater. In fact, ‘resonance (in) the heaven’ as a circumlocution for ‘thunder’ would point in the same direction. Linguistically, this reading would firmly assure the Judaean (vs. Phoenician, Israelite or Moabite) provenance of the text, as it is only in Southern Palestine that *ay was not contracted148. The quite uncommon shape of the hypothetic mem must advise much caution at this point, however.
147. Smith, Pitard 2009, 671.

148. Garr 1985, 35–40.
146 If one ventures to interpret no. 8 as a mirrored ṣade – what it looks like the most – the emerging sequence YṢʔ (no. 7–9) can scarcely be separated from Hebrew yāṣā(ʔ) ‘to go out’ and its pan-Semitic cognates. The following string PKBHM (no. 10–12) can then be divided into PK ‘your mouth’ (= Hebrew pīka) and BHM ‘in them’ (= Hebrew bāhäm). This is an appealing possibility in view of the fact that there is a steady connection between ‘mouth’ and ‘to go out’ in Semitic: Ugaritic b ph rgm l yṣʔ ‘The word has not yet gone out from his mouth’ (KAI 1.19 ii 26), Akkadian ina pī waṣû (CAD A2 371), Hebrew yāṣā(ʔ) dābār mi-ppä (BDB 423). This reading implies that instead of HŠMYM we must read HŠM (no. 4–6), equivalent to Hebrew ha-ššēm ‘the name’.
147 The final part of the inscription (no. 31–38) reads YRYNKY?N. This is a hard nut to crack, not the least because of the uncertain identification of the penultimate grapheme.
148 The former two or three letters are likely to represent a doubly weak root *yry (< *wry). As long as the preceding element is read as sk ‘pour out!’, an appealing possibility is *yāriyu ‘early rain’, reliably attested in Hebrew and Ugaritic as yōrä and yr respectively (BDB 435, DUL 961). The following N (no. 34) could then be interpreted as the 1 pl. pronominal suffix -nū: ‘our rain’ = ‘rain for us’.
149 As long as no. 37 is paleographically unclear, there can hardly be any reliable clue to KY?N. If read as kap, YKN could well represent *yakūn(u), a 3 sg. m. prefix conjugation from *kwn ‘to be firm, reliable’ or just ‘to be’. The preceding K would be equivalent to kī or kō.
150 With all due caution, the following tentative reading of the inscription can be proposed:
151 zdt hšm yṣʔ pk bhm ln ḥz bh wydbr k ʕl sk yryn k y?n
152 Translation. Resonance (of) the name. Your mouth (scil. ‘speech, command’) went out in them for us. He looked at it and said thus: “Come up! Pour down our rain! ...”
153 5. Some preliminary remarks on the inscription’s authenticity
154 There is no need to stress that any object promising a certain cultural-historical value and, incidentally, found outside regular excavations is suspect to be a forgery. Our squeeze is no exception, all the more since the original artefact is, in all probability, lost and the circumstances in which the copy was made are not known to us. The suspicions can undoubtedly be strengthened by an entire constellation of unusual features characterizing the inscription: a rare carving technique; the presence of ligatures149; uncertain genre; difficulties in interpretation. In brief, the risk of forgery is by no means insignificant.
149. A feature well known to be indicative of fakes since the Moabitica times (Heide 2012, 205–206).
155 These legitimate concerns can be counter-balanced by the following considerations in favor of the inscription’s authenticity.
156 To begin with, forgers usually fabricate their concoctions for one of the two principal purposes (which can, of course, be nicely combined): money and fame150. The second is not in issue here: the squeeze (let alone the original monument) has never been published and its very existence has been unknown. No scholar or dealer has ever profited from it to enhance his authority and prestige. It is similarly unclear what kind of material gain the hypothetic forger could obtain: the squeeze as such has no value whatsoever151, and if the original artefact were ever sold and bought, such a striking event could hardly have gone unnoticed, not the least because of the rather substantial costs such a deal would involve.
150. Cf. Rollston 2003, 191–192 for a more nuanced typology.

151. Indeed, to be best of our knowledge, forged estampages of NWS inscriptions have never featured in the authenticity debate.
157 A fake squeeze presupposes a fake object. In our case, it must have been a stone artefact of quite serious dimensions – something very far from a clay figurine or an ostracon. This would require a lot of time, energy, and artistic skills. One may doubt that such serious resources could be waisted by somebody who would not see any clear material and/or reputational reward for his work.
158 As persuasively shown by Chr. Rollston152, a modern forger has at his/her disposal an impressive array of technical and philological resources, which make him capable of producing objects nearly unassailable in both shape and contents. This was certainly not true, however, for an early, pre-WWI forger. True, the rather limited number of early NWS inscriptions known by that time could provide more or less reliable models for most of the signs featuring on the squeeze. A skillful potpourri of Mesha (1868), Zincirli (1893), Siloam (1882), Abibaʕal (1905), Zkr (1908) and Gezer (1909) is theoretically possible here, but it is an open question whether this was possible also in practice.
152. Rollston 2003, 136–139; 2004, 70–71.
159 If it was, the academic background of the forger must have been more than exemplary: the letters are produced with utmost care (nothing remotely similar to the cheap Moabitica-like fabrications) and, as argued above, the inscription does not seem to display any serious paleographic mixture, either chronological or geographical. The letters are generally archaic and compatible with a South Canaanite location, and there is hardly any serious deviation from this picture.
160 The inscription does not copy or imitate any known ancient text153. Yet, at the same time, the letters do not seem to follow each other at random: for quite a number of sequences, fairly promising NWS (notably, Hebrew) readings can be proposed and defended. In fact, there is hardly any group of letters which would look overtly unnatural by the established standards of NWS. No artificial dialectal mixture (e.g. Hebrew and Aramaic) is in evidence. The forger must have been a good philologist, particularly for his epoch.
153. For this feature characterizing many NWS forgeries see Rollston 2003, 142–150.
161 To recapitulate. The scholarly value of the squeeze is, in our view, fairly high, which has made its detailed publication an urgent task154. If eventually acknowledged as authentic155, its paramount relevance for the history of NWS scripts, languages and cultures is self-evident. If proved to be a fabrication, this fabrication will be quite unique in the history of our discipline and must have an exciting detective story behind it. Not a small revolution in the field of Semitic epigraphy – or in the field of Semitic epigraphic fakes. Time will tell.
154. In agreement with Heide’s judicious dictum: “Non-provenanced antiquities cannot be ignored; they must be published and assessed ... The best way to deal with unprovenanced artifacts is to publish them and to wait for comments, reviews, and additional publications where applicable” (2012, 231–232).

155. An in-depth laboratory analysis of the estampage’s paper is scheduled for a near future. It is expected to fix with enough precision the origin and age of the material. There is hope, moreover, that microparticles of the original stone could be detected.

Библиография

1. Abou-Assaf, A., Bordreuil, P., Millard, A. 1982: La statue de Tell Fekherye et son inscription bilingue assyro-araméenne. Paris.

2. Aharoni, Y. 1981: Arad Inscriptions. Jerusalem.

3. Aḥituv, Sh., Mazar, A. 2014: The inscriptions from Tel Reḥov and their contribution to the study of script and writing during Iron Age IIA. In: E. Eshel, Y. Levin (eds.), “See, I Will Bring a Scroll Recounting What Befell Me” (Ps 40:8). Epigraphy and Daily Life from the Bible to the Talmud Dedicated to the Memory of Professor Hanan Eshel. Göttingen, 39–68.

4. Al-Ghabban, A.I., André-Salvini, B., Demange, F., Juvin, C., Cotty, M. (eds.) 2010: Roads of Arabia. Archaeology and History of the Kingdom of Saudi Arabia. Paris.

5. Athas, G. 2003: The Tel Dan Inscription. A Reappraisal and a New Interpretation. Sheffield.

6. Avigad, N. 1953: The epitaph of a Royal Steward from Siloam Village. Israel Exploration Journal 3, 137–152.

7. Avigad, N., Sass, B. 1997: Corpus of West Semitic Stamp Seals. Jerusalem.

8. Balakhvantsev, A.S. 2016: The absolute chronology of Archaic Scythia: eastern benchmarks. In: A.S. Balakhvantsev, S.V. Kullanda (eds.), Kavkaz i step’ na rubezhe epokhi pozdney bronzy i rannego zheleza: materialy mezhdunzrodnoy konferentsii, posvyashchennoy pamyati Marii Nikolayevny Pogrebovoy [Caucasus and Steppes in the Late Bronze – Early Iron Age: Proceedings of the International Conference Dedicated to the Memory of Maria Nikolaevna Pogrebova]. Moscow, 15–24.

9. Балахванцев, А.С. Абсолютная хронология архаической Скифии: восточные реперы. В сб.: А.С. Балахванцев, С.В. Кулланда (ред.), Кавказ и степь на рубеже эпохи поздней бронзы и раннего железа: материалы международной научной конференции, посвященной памяти Марии Николаевны Погребовой. М., 15–24.

10. Basargina, Ye.Yu. 1995: [Russian Archaeological Institute in Constantinople: the archival collections]. In: I.P. Medvedev (ed.), Arkhivy russkikh vizantinistov v Sankt-Peterburge [Archives of Russian Byzantinists in Saint Petersburg]. Saint Petersburg, 62–92.

11. Басаргина, Е.Ю. Русский археологический институт в Константинополе: архивные фонды. В сб.: И.П. Медведев (ред.), Архивы русских византинистов в Санкт-Петербурге. СПб., 62–92.

12. Biran, A., Naveh, J. 1995: The Tel Dan inscription: a new fragment. Israel Exploration Journal 45, 1–18.

13. Bordreuil, P. 1977: Une inscription phénicienne champlevée des environs de Byblos. Semitica 27, 23–27.

14. Cross, F.M. 2003: Leaves from an Epigrapher’s Notebook: Collected Papers in Hebrew and West Semitic Palaeography and Epigraphy. Winona Lake.

15. Cross, F.M., McCarter, K.P. 1973: Two archaic inscriptions on clay objects from Byblos. Rivista di Studi Fenici 1, 3–8.

16. Dearman, J.A. (ed.) 1989: Studies in the Mesha Inscription and Moab. Atlanta.

17. Demsky, A. 1986: The ʻIzbet Ṣarṭah ostracon – ten years later. In: I. Finkelstein (ed.), ʻIzbet Ṣarṭah. An Early Iron Age Site near Rosh Haʻayin, Israel. Oxford, 186–197.

18. Deutsch, R., Heltzer, M. 1999: West Semitic Epigraphic News of the 1st Millennium BCE. Tel Aviv.

19. Deutsch, R., Lemaire, A. 2000: Biblical Period Personal Seals in the Shlomo Moussaieff Collection. Tel Aviv.

20. Dobbs-Allsopp, F., Roberts, J., Seow, C., Whitaker, R. 2005: Hebrew Inscriptions: Texts from the Biblical Period of the Monarchy with Concordance. New Haven–London.

21. Dupont-Sommer, A., Starcky, J. 1958: Les inscriptions araméennes de Sfiré (stèles I et II). Paris.

22. Eshel, E., Notarius, T., Dagan, A., Eniukhina, M., Workman, V., Maeir, A. 2022: Two Iron Age alphabetic inscriptions from Tell eṣ-Ṣâfi/Gath, Israel. Bulletin of the American Schools for the Oriental Research 388, online first (no page numbers).

23. Finkelstein, I., Sass, B., Singer-Avitz, L. 2008: Writing in Iron IIA Philistia in the light of the Tēl Zayit/Zētā abecedary. Zeitschrift des Deutschen Palästina-Vereins 124, 1–14.

24. Garr, R. 1985: Dialect Geography of Syria-Palestine. 1000–586 B.C.E. Winona Lake.

25. Gesenius, W. 1910: Hebrew Grammar. Oxford.

26. Gogel, S. 1988: A Grammar of Epigraphic Hebrew. Atlanta.

27. Hamilton, G. 2014: Reconceptualizing the periods of early alphabetic scripts. In: J.A. Hackett, W. Aufrecht (eds.), “An Eye for Form”: Epigraphic Essays in Honor of Frank Moore Cross. Winona Lake, 30–55.

28. Heide, M. 2012: The Moabitica and their aftermath. How to handle a forgery affair with an international impact. In: E. Lubetski, M. Lubetski (eds.), New Inscriptions and Seals Relating to the Biblical World. Atlanta, 193–241.

29. Herr, L. 2014: Aramaic and Ammonite seal scripts. In: J.A. Hackett, W. Aufrecht (eds.), “An Eye for Form”: Epigraphic Essays in Honor of Frank Moore Cross. Winona Lake, 175–186.

30. Lemaire, A., Sass, B. 2013: The mortuary stele with Sam’alian inscription from Ördekburnu near Zincirli. Bulletin of the American Schools for Oriental Research 369, 57–136.

31. Maeir, M., Wimmer, S., Zukerman, A., Demsky, A. 2008: A late Iron Age I / early Iron Age II Old Canaanite inscription from Tell eṣ-Ṣâfī/Gath, Israel: palaeography, dating, and historical-cultural significance. Bulletin of the American Schools of Oriental Research 351, 39–71.

32. Maeir, A., Eshel, E. 2014: Four short alphabetic inscriptions from Late Iron Age IIa Tell eṣ-Ṣafi/Gath and their implications for the development of literacy in Iron Age Philistial and Environs. In: E. Eshel, Y. Levin (eds.), “See, I Will Bring a Scroll Recounting What Befell Me” (Ps 40:8). Epigraphy and Daily Life from the Bible to the Talmud Dedicated to the Memory of Professor Hanan Eshel. Göttingen, 69–88, 205–210.

33. Mazar, A. 2003: Three 10th–9th century BCE inscriptions from Tel Rehov. In: C. den Hertog, U. Hübner, S. Münger (eds.), Saxa loquentur. Studien zur Archäologie Palästinas/Israels. Festschrift für Volkmar Fritz zum 65. Geburtstag. Münster, 171–184.

34. Mazar, E., Ben-Shlomo, D., Aḥituv, S. 2013: An inscribed pithos from the Ophel, Jerusalem. Israel Exploration Journal 63, 39–49.

35. McCarter, K.P. 1975: The Antiquity of the Greek Alphabet and the Early Phoenician Scripts. Missoula.

36. McCarter, K.P. 2008: Paleographic notes on the Tel Zayit abecedary. In: R. Tappy, P. K. McCarter (eds.), Literate Culture and Tenth-Century Canaan: The Tel Zayit Abecedary in Context. Winona Lake, 45–59.

37. McCarter, K.P., Bunimovitz, S., Lederman, Z. 2011: An archaic Baal inscription from Tel Beth-Shemesh. Tel Aviv 38, 179–193.

38. Meshel, Z. 2012: Kuntillet ʕAjrud (Ḥorvat Teman). An Iron Age II Religious Site on the Judah-Sinai Border. Jerusalem.

39. Millard, A. 1970: “Scriptio Continua” in early Hebrew: ancient practice or modern surmise? Journal of Semitic Studies 15, 2–15.

40. Möller, G. 1909: Hieratische Paläographie. Bd. I. Leipzig.

41. Naveh, J. 1973: Word division in West Semitic writing. Israel Exploration Journal 23, 206–208.

42. Naveh, J. 1982: Early History of the Alphabet. An Introduction to West Semitic Epigraphy and Palaeography. Jerusalem.

43. Osborne, J. 2020: The Syro-Anatolian City-States: An Iron Age Culture. New York.

44. Pardee, D. 2009: A new Aramaic inscription from Zincirli. Bulletin of the American Schools for Oriental Research 356, 51–71.

45. Pavlichenko, N.A. 2018: The squeezes of Greek inscriptions in the collection of the Russian Archaeological Institute in Constantinople]. In: I.V. Tunkina (ed.), Millerovskiye chteniya–2018: Preyemstvennost i traditsii v sokhranenii i izuchenii dokumentalnogo akademicheskogo naslediya. Materialy II Mezhdunarodnoy nauchnoy konferentsii 24–26 maya 2018 g., Sankt-Peterburg [Müller Conference 2018. Continuity and Traditions in Preserving and Studying Documentary Academic Heritage. Materials of the 2nd International Scientific Conference. May 24–26, 2018, St. Petersburg]. Saint Petersburg, 465–473.

46. Павличенко, Н.А. Коллекция эстампажей греческих надписей в фонде Русского Археологического института в Константинополе. В сб.: И.В. Тункина (ред.), Миллеровские чтения–2018. Преемственность и традиции в сохранении и изучении документального академического наследия. Материалы II Международной научной конференции 24–26 мая 2018 г., Санкт-Петербург. СПб., 465–473.

47. Pilkington, N. 2019: The Carthaginian Empire: 550–202 BCE. New York.

48. Renz, J. 1995a: Die althebräischen Inschriften. Teil 1. Text und Kommentar (Handbuch der althebräischen Epigraphik, I). Darmstadt.

49. Renz, J. 1995b: Die althebräischen Inschriften. Texte und Tafeln (Handbuch der althebräischen Epigraphik, III). Darmstadt.

50. Renz, J. 1995c: Die althebräischen Inschriften. Teil 2. Zusammenfassende Erörtungen, Paläographie und Glossar (Handbuch der althebräischen Epigraphik, II/1). Darmstadt.

51. Roche, M.-J. 2020: Inscriptions en champlevé de Taymā’. Semitica 62, 163–205.

52. Rollston, C. 2003: Non-provenanced epigraphs I: pillaged antiquities, Northwest Semitic forgeries, and protocols for laboratory tests. Maarav 10, 135–193.

53. Rollston, C. 2004: Non-provenanced epigraphs II: the status of non-provenanced epigraphs within the broader corpus of Northwest Semitic. Maarav 11, 57–79.

54. Rollston, C. 2008: The Phoenician script of the Tel Zayit abecedary and putative evidence for Israelite literacy. In: R. Tappy, P.K. McCarter (eds.), Literate Culture and Tenth-Century Canaan: The Tel Zayit Abecedary in Context. Winona Lake, 61–96.

55. Rollston, C. 2014a: The Iron Age Phoenician script. In: J.A. Hackett, W. Aufrecht (eds.), “An Eye for Form”: Epigraphic Essays in Honor of Frank Moore Cross. Winona Lake, 72–99.

56. Rollston, C. 2014b: Northwest Semitic cursive scripts of Iron II. In: J.A. Hackett, W. Aufrecht (eds.), “An Eye for Form”: Epigraphic Essays in Honor of Frank Moore Cross. Winona Lake, 202–234.

57. Sass, B. 2017: The emergence of monumental West Semitic alphabetic writing with an emphasis on Byblos. Semitica 59, 109–141.

58. Sass, B., Garfinkel, Y., Hasel, M., Klingbeil, M. 2015:The Lachish jar sherd: an early alphabetic inscription discovered in 2014. Bulletin of the American Schools for the Oriental Research 374, 233–245.

59. Schmitz, P.C. 2008: Deity and royalty in dedicatory formulae: the Ekron store-jar inscription viewed in the light of Judges 7:18, 20 and the inscribed gold medallion from the Douïmès necropolis at Carthage (KAI 73). Maarav 15/2, 165–173.

60. Schmitz, P.C. 2014: Phoenician seal script. In: J.A. Hackett, W. Aufrecht (eds.), “An Eye for Form”: Epigraphic Essays in Honor of Frank Moore Cross. Winona Lake, 141–174.

61. Schniedewind, W. 2005: Problems in the paleographic dating of inscriptions. In: T. Levy, T. Higham (eds.), The Bible and Radiocarbon Dating: Archaeology, Text and Science. Oakville, 405–412.

62. Sebbane, M. 2016. ḥnn gaming board: two-sided gaming board fragment bearing an ownership inscription. In: S. Bunimovitz, Z. Lederman (eds.), Tel Beth-Shemesh: A Border Community in Judah: Renewed excavations 1990–2000: The Iron Age. Tel Aviv, 639–646.

63. Smith, M., Pitard, W. 2009: The Ugaritic Baal Cycle. Vol. II. Introduction with Text, Translation and Commentary of KTU/CAT 1.3–1.4. Leiden.

64. Stager, L. 1969: An inscribed potsherd from the eleventh century. Bulletin of the American Schools of Oriental Research 194, 45–52.

65. Tappy, R., McCarter, P.K., Lundberg, M., Zuckerman, B. 2006: An abecedary of the mid-tenth century from the Judean Shephelah. Bulletin of the American Schools for Oriental Research 344, 5–46.

66. Tropper, J. 1993: Die Inschriften von Zincirli. Neue Edition und vergleichende Grammatik des phönizischen, samʾalischen und aramäischen Textkorpus. Münster.

67. Tropper, J. 2012: Ugaritische Grammatik. Münster:

68. Tunkina, I. V., Bondar, L. D., Ponikarovskaya, M. V., Pavlichenko, N. A., Chernova N. V. 2020: [The squeezes of the Russian Archaeological Institute in Constantinople (1895–1914) in the collection of the St. Petersburg Branch of the Archive of the Russian Academy of Sciences: scientific restoration, comprehensive study, creation of high-precision digital 3D models]. Peterburgskiy istoricheskiy zhurnal [Saint-Petersburg Historical Journal] 4 (28), 194–203.

69. Тункина И. В., Бондарь Л. Д., Поникаровская М. В., Павличенко Н. А., Чернова Н.В. Эстампажи Русского археологического института в Константинополе (1895–1914) в собрании Санкт-Петербургского филиала Архива РАН: Научная реставрация, комплексное изучение, создание высокоточных цифровых 3D-моделей. Петербургский исторический журнал. 2020. № 4 (28). С. 194–203.

70. Vanderhooft, D. 2014: Iron Age Moabite, Hebrew, and Edomite monumental scripts. In: J.A. Hackett, W. Aufrecht (eds.), “An Eye for Form”: Epigraphic Essays in Honor of Frank Moore Cross. Winona Lake, 107–126.

71. Wimmer, S. 2008: Palästinisches Hieratisch: die Zahl- und Sonderzeichen in der althebräischen Schrift. Wiesbaden.

Комментарии

Сообщения не найдены

Написать отзыв
Перевести