THE Old Testament--except, possibly, the latest portions--was not written originally in the characters with which we are familiar; and a recollection of the change through which the Hebrew alphabet passed is preserved both in the Talmud and by the Fathers. In the Talmud, Sanh. 21ᵇ, we read: 'Originally the law was given to Israel in the Hebrew character and in the sacred tongue: it was given again to them, in the days of Ezra, in the "Assyrian" character (בכתב אשורי), and in the Aramaic tongue. Israel chose for themselves the "Assyrian" character and the sacred tongue, and left to the ἰδιῶται the Hebrew character and the Aramaic tongue. Who are the ἰδιῶται? R. Ḥasda¹ said, The Cuthites [i.e. the Samaritans: 2 Ki. 17, 24]. What is the Hebrew character? R. Hasda said, ³ ²כתב ליבונאה.' The original character is here termed Hebrew (כְתָב עִבְרִ), the new character אשורי ⁴. In the Jerus. Talmud, Megillah 1, 71ᵇ, two explanations are offered of the latter term: 'And why is it called אשורי? Because it is straight (מְאֻשָׁר) in form. R. Levi says, Because the Jews brought it home with them from Assyria⁵. The explanation Assyrian is

[Footnotes]

¹ A teacher of the school of Sura, d. 309.

² בתחלה ניתנה תורה לישראל בכתב עברי ולשון הקודש חזרה וניתנה להם בימי עזרא בכתב אשורית ולשון ארמי ובירדי להן לישראל כתב אשורית ולשון הקודש והניחו להדיוטות כתב עברי ולשון ארמית מאן הדיוטות אמר ר׳ חסדא כותאי מאי עברית אמר ר׳ חסדא כתב ליבונאה.

³ An expression of uncertain meaning: comp. Hoffmann in the ZATW. i. 337; Levy NHWB. s. v.

⁴ The same term is used elsewhere: thus in the Mishnah, Megillah 1, 8 אין בין ספרים לתפלין ומזוזות אלא שהספרים נכתין בכל לשון ותפלין ומזוזות אינן נכתין אלא אשורית i.e. the sacred books might be written in any language, but the Tefillin and Mezuzoth only in the 'Assyrian' character.

⁵ ולמה נקרה שמו אשורי שהוא מאושר בכתבו אמר ר׳ לוי על שם שֶׁעָלָה בְיָדָם מאשור.

1365 b

[page ii -- Introduction]

the more probable, whether it be supposed to be used loosely for 'Babylonian,' or whether--as others have thought--it have the sense of Syrian or Aramaic (as occasionally in later times appears to have been the case¹), and so embody a true tradition as to the origin of the new character. The כתב אשורי is that which in later times acquired the name of כְתָב מְרֻבָּע or square character². Origen, speaking of the sacred name, says that in accurate MSS. it was written in archaic characters, unlike those in use in his own day³: ἔστι δὲ παῤ αὐτοῖς καὶ τὸ ἀρχιέρεως ἐγέγραπτο· κύριος δὲ καὶ τοῦτο παῤ Ἔλλησι ἐκφωνεῖται. Καὶ ἐν τοῖς ἀκριβέσι τῶν ἀντιγράφων Ἑβραικοῖς ἀρχαίοις γράμμασι γέγραπται ἀλλ’ οὐχὶ τοῖς νῦν. Φασὶ γὰρ τὸν Ἔσδραν ἑτέροις γρήσασθαι μετὰ τὴν αἰχμαλωσίαν. In his Commentary on Ez. 9, 4 he adds that a converted Jew, in answer to an enquiry, told him that τὰ ἀρχαῖα στοιχεῖα ἐμφερὲς ἔχειν τὸ θαῦ τῷ τοῦ σταυροῦ χαρακτῆρι. Jerome, at the beginning of the 'Prologus Galeatus⁴,' after observing that the Hebrews, Syrians, and Chaldaeans had all an alphabet of twenty-two characters, continues, 'Samaritani etiam Pentateuchum Moysi totidem litteris scriptitant, figuris tantum et apicibus discrepantes. Certumque est Esdram scribam legisque doctorem, post capta Hierosolyma et instaurationem templi sub Zorobabel, alias litteras repperisse quibus nunc utimur, cum ad illud usque tempus iidem Samaritanorum et Hebraeorum characteres fuerint.' On Ez. 9, 4 he makes a remark to the same effect as Origen. In his letter to Marcella, De decem nominibus Dei⁵, he writes, _Nomen τετραγράμματον quod ἀνεκφωνητον id est ineffabile putaverunt quod his litteris scribitur יהוה: quod quidam non intelligentes propter elementorum similitudinem cum in Graecis

[Footnotes]

¹ Cf. Jer. 35 (42), 11. Ez. 32, 29 (Ἀσσύριοι for אדמ, i.e. ארמ) in the LXX.
² For other statements made by the Jews respecting the change of script, and often dependent upon most fanciful exegesis, see Chapman, Introd. to the Pentateuch (uniform with the Cambridge Bible), 1911, pp. 279-287).
³ On ψ. 2, 2 (quoted by Montfaucon, Hexapla, i. 86: in a slightly different form, from other MSS., in ed. Bened. ii. 539 = Lommatzsch xi. 396 f.).
⁴ Or Preface to the Four Books of Kings (which were the first translated by Jerome from the Hebrew), designed as a defence (galea) against detractors,--printed at the beginning of ordinary editions of the Vulgate.
⁵ Ep. 25 (ed. Bened. i. 705; Vallarsi i. 129).

[page iii -- § 1. _Change of Character in the Hebrew Script]

litteris repererent ΠΙΠΙ legere consueverunt¹.' Epiphanius² (d. 403) makes a statement similar to that contained in the extract from Sanhedrin, that a change of character was introduced by Ezra, and that the old form was only retained by the Samaritans.

The fact of a change of character, to which these passages bear witness, is correct: the only error is that it is represented as having been introduced by one man. Tradition, as is its wont, has attributed to a single age, and to a single name, what was in reality only accomplished gradually, and certainly was not completed at the time of Ezra (who came to Palestine B.C. 458).

What, then, was that older character of which the Talmud and the Fathers speak, and which they describe as being still retained by the Samaritans? It was the character which, with slight modifications of form, is found upon the Inscription of Mesha‘ (commonly known as the 'Moabite Stone'), upon early Aramaic and Hebrew gems, upon Phoenician Inscriptions, and upon the few early Hebrew Inscriptions which we at present possess, viz. those found at Samaria, Gezer, and Siloam³. It was the common Semitic character, used alike, in ancient times, by the Moabites, Hebrews, Aramaeans, and Phoenicians, and transmitted by the Phoenicians to the Greeks. This character remained longest without substantial alteration in Hebrew proper and Phoenician: in Greek it changed gradually to the character with which we are now familiar: the transition to what is termed above the כתב אשורי was effected first in Aramaic; it was only accomplished at a later period in Hebrew, in consequence, no doubt, of the growing influence of the Aramaic language in Palestine, in the period immediately preceding the Christian era.

Tables of the chief ancient Semitic alphabets are to be found in

[Footnotes]

¹ Comp. the Hexapla on ψ. 26 (25), 1; Is. 1, 2 (with Dr. Field's note); Nestle in the ZDMG. xxxii. 466-9, 507.

In the palimpsest Fragments of the Books of Kings [1 Ki. 20, 7-17; 2 Ki. 23, 11-27] in Aquila's Translation, found by Dr. Schechter in the Cairo Genizah, and published by F. C. Burkitt in 1897, and in those from the Psalms, published in C. Taylor's Cairo Genizah Palimpsests (1900), the Tetragrammaton is regularly written in the archaic characters here referred to (cf. Burkitt, p. 15 f.; DB. iv. 444).

² De xii gemmis, § 63 (ed. Dindorf, 1863, IV. 213; cited by Hoffmann, u.s. p. 334).

³ See p. vii ff.

b 2

[page iv -- 1. Introduction ]

most Hebrew grammars of modern times¹, and they need not be here repeated. It will be more instructive to place before the reader specimens of Inscriptions themselves in facsimile. The earliest Inscription of all, that of Mesha‘ (c. B.C. 900), has not been included, as facsimiles of it with transcriptions in modern Hebrew characters are readily obtainable². The characters used in this Inscription are the most ancient of the West-Semitic type that are known³, though they differ but slightly from the earliest of those that are figured below: the differences may be studied in detail with the aid of the Tables mentioned below.

Here are examples of seals with Aramaic (Figs. 1 and 2) and Hebrew (Figs. 3 and 4) Inscriptions, the first three of which are

assigned by M. A. Levy⁴ to the eighth cent. B.C., while the fourth is somewhat later.

[Footnotes]

¹ There is a good one at the beginning of Gesenius-Kautzsch. More extensive Tables may be found in Cooke's North-Semitic Inscriptions (1903), Plates XII-XIV; in Plates XLIV-XLVI of the Atlas to Lidzbarski's Handbuch der nordsemitischen Epigraphik (1898); and especially in Chwolson's Corpus Inscriptionum Hebraicarum enthaltend Grabinschriften aus der Krim, etc., 1882 (a Table constructed by the eminent German palaeographer Euting, containing specimens of not less than 139 alphabets).

² See Die Inschrift des Königs Mesa von Moab für akademische Vorlesungen herausgegeben von Rudolf Smend und Albert Socin (Freiburg i. B., 1886); and Plate I in Lidzbarski's Handbuch (above, n. 1).

³ The Inscription on fragments of a bowl dedicated to בעל לבנן, found in Cyprus in 1872, is, however, considered by some to be of greater antiquity (see Cooke, NSI. No. 11). The characters are very similar (Lidzb. Atlas, II. ī).

⁴ Siegel und Gemmen mil aramäischen, phönizischen, althebräischen etc. Inschriften (Breslau, 1869), pp. 6, 8, 34, 37.

[page v -- § 1. Old West-Semitic and Greek Inscriptions ]

No. 1 was found under the pedestal of a colossal bull at Khorsabad: Nos. 3 and 4 were obtained by M. Waddington, the former in Aleppo, the latter in Damascus. The resemblance of some of the characters to those of the Greek alphabet will be evident: the ד and ס are closely similar to Δ¹ and Ξ, while the forms of ה and ר become, when turned round so as to face the right, Ε and Ρ respectively. The ל and ע exhibit quite the forms which they still have in modern European alphabets, L and O, but from which in the later Hebrew alphabet they both diverged considerably. The characters on old Phoenician seals and gems are so similar that it has not been deemed necessary to add illustrations². The following specimens of ancient Inscriptions from Thera will illustrate the derivation of the Greek alphabet from the Phoenician³: the letters, as is often the case in the most ancient Greek Inscriptions, are read from right to left:

The Ε does not differ materially from the ה in Fig. 3; the Π differs but slightly from the פ of Mesha‘'s Inscription, and indeed agrees

[Footnotes]

¹ In the Inscription of Mesha‘, as in that to בעל לבנן, from Cyprus (Cooke, NSI. No. 11; Lidzb., Plate II, A), the ד is a simple triangle, with no elongation of the right side downwards; it thus exactly resembles the Greek Δ, and is also distinct from the ר.

² Examples may be seen in Levy, l. c. Taf. II; cf. Cooke, Pl. IX, B 1-7.

³ For two other rather interesting examples, from the Gortynian Code, and the Treaty between the Eleans and the Heraeans (c. 525 B.C.), see Berger, Hist. de l'Écriture dans l'Antiquité² (1892), pp. 132-4 (also in Roberts, Greek Epigraphy, Pt. i. (1887), pp. 42, 288,--with many other facsimiles of archaic Greek inscriptions, pp. 23 ff., 39 ff., etc.).

[page vi -- Introduction ]

substantially with the ף of modern printed texts : the Γ and Κ are quite the ג and כ of Mesha‘'s: the Ι, which has not yet become a straight line, retains evident traces of its origin (cf. Fig. 3): the Μ as compared with the Ν has a double turn at the top, exactly as in Fig. 3, the Ρ and the Δ are more differentiated, but do not differ in principle from the forms in Figs. 1 and 2. By turning the letters round so as to face the right, the later and usual form of the Greek character is (in most cases) immediately produced. The evidence of Inscriptions thus confirms the testimony of Herodotus, respecting the origin of the Greek alphabet from Phoenicia¹.

The most ancient West-Semitic Inscriptions, at present known, next to that of Mesha‘, are probably the בעל לבנן Inscription from Cyprus (p. iv n. 3), and the Old Aramaic Inscriptions of Zinjirli, near

[Footnotes]

¹ Hd. 5. 58 Οἱ δὲ Φοίνικες οὗτοι οἱ σύν Κάδμῳ ἀπικόμενοι ... ἄλλα τε πολλά, οἰκήσαντες ταύτην τὴν χώρηω, ἐσήγαγον διδασκάλια ἐς τοὺς Ἔλληνας, καὶ δὴ καὶ γράμματα, οὐκ ἐόντα πρὶν τοῖς Ἕλλησι, ὡς ἐμοὶ δοκέεν· πρῶτα μέν, τοῖσι καὶ ἅπαντες χρέωνται Φοίνικες· μετὰ δέ, χρόνου προβαίνοντος, ἅμα τῇ φωνῇ μετέβαλον καὶ τὸν ῥυθμὸν (the shape) τῶν γραμμάτων. Περιοίκεον δέ σφεας τὰ πολλὰ τῶν χώρων τοῦτον τὸν χρὸνον Ἑλλήνων Ἴωνες. οἳ παραλαβὸντεσ διδαχῇ παρὰ τῶν φοινίκων τὰ γράμματα μεταρρυθμίσαντές σφεων ὀλίγα ἐχρέωντο. Archaic Greek characters are termed by him accordingly (ib. 59) Καδμέϊα γράμματα.

A little consideration will shew generally, how by continued modification in different directions, the Greek and modern European character on the one hand, and the Hebrew square character on the other, have been developed from a common origin. Out of the archaic ב, the Greek Β arose by turning the letter from left to right, and carrying round the lower part of it so as to form a complete semicircle: the square ב arose by the opening and ultimate disappearance of the upper part of the original letter, as explained below (p. xiv f.). Δ and Ρ in Greek preserved the distinctness of type which these letters shew on Mesha‘'s Inscription: by the addition of a tail to the ד, and the gradual degeneration of the upper part of both letters, they acquired the great similarity of form which they present in most of the later Hebrew alphabets. Eshmun‘azar's ז is almost our Z; by successive shortening of the strokes, and extension of the angles between them, ז is produced. The old ל is nearly our L: by the addition of a tail on the right, the square ל is produced. Mesha‘'s ע is our O; the first stage in the derivation of ע will appear in Plate III. Out of the old ף, the Greek Π arose by the gradual prolongation downwards of the upper left-hand part of the letter (see the first stage in Fig. 5): the final ף is nearly the same as the old form; the medial פ merely differs from it by the turn to the left given to the lower part of the letter, when the end of a word did not bring the scribe's hand to a pause (cf. p. xix). The crooked Ι of the archaic Greek (Fig. 5; Roberts, 23 ff., 40 ff.) before long becomes straight (ib. 30, 61).

[Blank Page]

[page vii -- § 1. The Gezer Inscription ]

Aleppo (8th cent. B.C.)¹. For our present purpose, however, these may be passed by; and we may look at what is at present the most ancient Hebrew Inscription known, the Calendar-Inscription discovered in 1908 at Gezer (Plate I)². Its date is uncertain, but in any case it is later than Mesha‘'s Inscription, and earlier than the Siloam Inscription (p. ix). Those who think that the Siloam Inscription is not earlier than the 3rd cent. B.C., place it in the 6th cent. B.C.³; Lidzbarski considers it 'much older than the 6th century⁴; and G. B. Gray assigns it to the 8th century⁵.

The Inscription reads (Lidzbarski)--

I.e. 1 The month of ingathering [Tishri]. The month of (2) sowing. The month of late sowing. 3 The month of cutting (or hoeing up?) flax. 4 The month of barley-harvest. 5 The month of the general harvest. 6 The month of (vine-)pruning. 7 The month of summer-fruits.

⁦1. ירחו. Though ירח ואסף might be read (and similarly in the following lines),⁩ 'A month and ingathering' yields a poor sense; and it seems that, in spite of its rarity in the OT. (only once in prose, Gen. 1, 24 חיתו ארץ), the ו is the old case-ending, the 12 occurrences of which in OT. are given in GK. § 90⁰. Was this of more frequent occurrence in the autographs of the OT. than it is in

[Footnotes]

¹ See Cooke, NSI. p. 159ff.; and, for the characters, the Atlas to Lidzbarski's Handbuch, Plates XXII-XXIV, XLV, col. 1.

² The inscriptions on ostraka, found in 1910 on the site of the ancient Samaria, and belonging to the time of Ahab (PEFQS. 1911, p. 79 ff.), are more ancient; but facsimiles of these are not at present (July, 1912) available.

³ Stanley A. Cook, PEFQS. 1909, p. 308 f.

⁴ Ibid. p. 26; Ephemeris, iii. 37.

⁵ PEFQS. 1909, p. 32.

[page viii -- Introduction ]

⁨MT.? אָסִף Ex. 23, 16 וְחַג הָאָסִף בצאת בְּאָסִףךָ את־מעשיך מן השדה⁦. 34, 22†.   2.⁩⁩ לֶקֶש (Am. 7, 1†, differently), or (Marti, p. 225) לָקִש , here, apparently, the 'late' sowing in Feb. (Dalman, PEFQS. 1909, 1909, p. 118; cf. Wetzstein, ap. Delitzsch on Job 24, 6).   3. עֶצֶד or (or עֲצֹד) cf. מַעֲצָד Is. 44, 12. Jer 10, 3 (an axe for cutting trees). In Ethiopic עצד is to reap. Flex is usually pulled up; but it may have been anciently cut in Palestine, as it still is about Aleppo (ibid. p. 90). Or (Dalm.) it may have been cut out of the ground with a מַעֲצָד, as a קַרְדֹּם was used in time of harvest (Pē’āh iv. 4). פֶּשֶׁת, cf. פִּשׁתִּי Hos. 2, 7. The month meant is March.   4. קְצִר שְׂעֹרִם (2 Sam. 21, 9), in April. The ס is placed below the line for want of space.   5. The month of the reaping (or harvest) of all things, i.e. of the general harvest in May.   6. The pruning (זָמִר Ct. 2, 12) meant will be (Dalm. p. 119), the second pruning, in June.   7. קץ (i.e. קָיִץ) the late summer fruits (see on 2 Sam. 16, 1), ripe in July or August. The Calendar is imperfect, containing only 8 months: but this and other difficulties connected with it need not here be considered¹.

The characters are bold and clear, though evidently the work of an unpractised hand. Most of the characters have archaic forms (compare, for instance, the א, ד, ו, ז, ח, ס, צ, ק, ש with the earlier

The characters are bold and clear, though evidently the work of an unpractised hand. Most of the characters have archaic forms (compare, for instance, ‭the א, ד, ו, ז, ח, ס, צ, ק, ש with‬ the earlier forms in the Tables of Cooke, Lidzbarski, or GK.): there are few or none of the curves, or other modifications, which are characteristic of the later forms. The ב in l. 5 is very abnormal; but this may be due to the inexperience of the engraver. The letters at the lower left-hand corner are read by Lidzbarski ‭as . . . . .‮אב‬,--perhaps ‮אֲבִיֿצָֿ[דָק]‬ ².‬

Until the discovery of the Gezer Inscription, the Inscription on the wall of the tunnel of Siloam (Plate II) was considered to be the oldest known Hebrew Inscription. The Pool of Siloam is situated at the extreme S. of the Eastern hill of Jerusalem (on the N. of which the Temple formerly stood), near the entrance to the Tyropoeon valley; and a conduit or tunnel cut through the rock from the Virgin's

[Footnotes]

¹ See further PEFQS. 1909, 26 ff. (Lidzbarski), 30 ff. (G. B. Gray), 113 ff. (Daiches, on Babylonian parallels), 118 f. (Dalman), 189 ff. (Gray), 194 f. (Lidzbarski); Lidzbarski's Ephemeris, iii. 37 ff. (notice, p. 45, the parallel from Tosefta, p. 215, l. 15 ff., ed. Zuckermandel); Marti, ZAW. 1909, p. 222 ff.

² The line above a letter indicates that the reading is not quite certain.

[Blank page]

[page ix -- § 1. The Siloam Inscription ]

Spring¹--the one natural spring which Jerusalem possesses--situated some distance above it, on the E. side of the same hill, leads down to it, and supplies it with water². The tunnel is circuitous, measuring 1708 feet (Warren), or 1757 feet (Conder), though the distance in a straight line is considerably less. At a distance of about 19 feet from where the tunnel opens into the Pool of Siloam, and on the right-hand side as one enters it, is an artificial niche or tablet in the rock, the lower part of which is occupied by the Inscription. The Inscription was first observed in 1880, by a pupil of Architect Schick, who, while wading in the Pool with a lighted candle, observed what appeared to be characters engraved on the rock. Ultimately, in 1881, a gypsum cast was obtained by Dr. Guthe, who published a photograph, with accompanying description, in 1882³, which has since been often reproduced. A portion of three lines in the Inscription has been destroyed through the wearing away of the rock; but the general sense is quite plain. Here is the Inscription, transliterated into modern Hebrew characters:

I.e. 1. [Behold] the piercing through! And this was the manner of the piercing through. Whilst yet [the miners were lifting up]

2. the pick, each towards his fellow, and whilst yet there were three cubits to be pierced [through, there was heard] the voice of each call-

3. ing to his fellow, for there was a fissure (?) in the rock on the right-hand . . . . . . . . And on the day of the

[Footnotes]

¹ Not the Virgin's Pool, as stated incorrectly in the Palaeographical Society's Volume. This is a small artificial reservoir near St. Stephen's Gate, and has no connexion with either the Virgin's Spring, or the Pool of Siloam.

² See the Plan in EB. ii, facing col. 2419-20, or G. A. Smith, Jerusalem (1907), ii, Plan facing p. 39; and comp. i. 87-92.

³ ZDMG. 1882, pp. 725-50. See also Lidzbarski, Ephemeris, i. 53.

[page x -- Introduction ]

4. piercing through, the miners (lit. hewers) smote each so as to meet his fellow, pick against pick: and there flowed

5. the water from the source to the pool, 1200 cubits; and one hun-

6. dred cubits was the height of the rock over the head of the miners.

The Hebrew is as idiomatic, and flowing, as a passage from the Old Testament.   1. נִקְבָּה or נָקְבָּה does not occur in the OT.: נקב is to pierce (2 Ki. 12, 10 al.); ܢܶܩܒܳܐ is a hole or aperture.--On the use of דבר, comp. p. 192 note.   2. רֵעוֺ as Jer. 6, 21: ‭usually ‮רֵעֵהוּ‬.--‮בעוד‬ as‬ Gen. 48, 7, cf. Am. 4, 7.   3. חית, i.e. probably חָיָת as 2 Ki. 9, 37 Kt.--זדה: the letters are quite clear, but the meaning is altogether uncertain, the word being not otherwise known, and the derivation from זוּד producing no suitable sense.   4. לקרת, vocalize לִקרֹת, the infin. ‭of ‮קָרָה‬   5. The‬ order of the numerals in מאתים ואלף (the smaller before the greater), as Nu. 3, 50 שלש מאות ואלף; but the order is rare in OT., except in P, Ez. Chr. (GK. § 134ⁱ), and with אלף very rare¹.   5-6. מְאַת אמה, as מְאַת שנה Gen. 5, 3, and often besides in P (LOT. p. 131 (edd. 1-5, p. 124), No. 8; GK. § 134ᵍ). On the orthography of the Inscription, see below, pp. xxx, xxxii. The words, as in the Inscription of Mesha‘, are separated by dots, without spaces².

The Inscription has been generally assigned to the time of Hezekiah, who is stated to have 'made the pool, and the conduit, and brought water into the city' (2 Ki. 20, 20) 'to the west side of the city of David' (2 Ch. 32, 30) in terms which appear exactly to describe the function of the tunnel in which the Inscription is³.

E. J. Pilcher, however (PSBA. 1897, p. l65 ff., with a Table of Alphabets; 1898, p. 213 ff.), pointed out the resemblance of several of its characters to those of a later date, and argued that it belonged to the time of Herod. His conclusions were combated by Conder (PEFQS. 1897, p. 204 ff.): he replied ibid. 1898, p. 56 f. Stanley A. Cook, in his detailed palaeographical study of the Old Hebrew alphabet in the PEFQS. 1909, p. 284 ff., though not accepting a date as late as this, agrees (cf. p. 305 bottom) that the characters point to a date later than c. 700 B.C.: 'if placed early,' he remarks (p. 308), 'it embarrasses, and will always embarrass, Hebrew palaeography;' he cannot, indeed (ibid. n. 2), fix the approximate date with any confidence, but thinks a date in the time of Simon, son of Onias (see Ecclus. 50, 3 Heb.),--probably c. 220 B.C.,--not impossible. Let us hope that future discoveries will make the date clearer.

[Footnotes]

¹ Add 1 Ki. 5, 12, Ez. 48, 16. 30. 32. 33. 34; and see, for further particulars, Herner, Syntax der Zahlwörter im AT., 1893, pp. 72 f., 74, 79.

² See further, NSI. No. 2.

³ Guthe, l. c. pp. 745-8; Smith, i. 102 f., ii. 151.

[page xi -- § 1. The Siloam Inscription ]

For our present purpose it is not necessary to consider this question further. Although some of the Siloam characters do resemble the later, rather than the earlier, examples of the older script (see, in Lidzbarski's Plate XLVI, Table III, the parallel cross strokes of the א, the ז, the curving tail ‭in כ, מ, נ, and פ, and‬ the disappearance of the left-hand upright stroke of the צ), they are still substantially of the archaic type, and there is no appreciable approximation to the square type.

The Samaritan character, as stated in the passages quoted above from the Talmud and the Fathers, preserves in all essential features the old Hebrew type, the modifications being confined to details, and originally, no doubt, being merely calligraphic variations:--

In Palestine the old Hebrew character was used regularly on coins, from the earliest Sheqels and half-Sheqels struck by Simon Maccabaeus (B.C. 141-135) to those of the Great Revolt, A.D. 65-68, and of Simon Bar-cochab, A.D. 132-135¹. The example (Fig. 7) is a Sheqel of the third year (ש ג i.e. שנה ג) of Simon Maccabaeus:--

As characters that were entirely unknown would evidently not be suitable for use upon coins, it may be inferred that though in the time of Christ the older character had been generally superseded (for the י, Matth. 5, 18, is by no means the smallest letter in the old alphabet), it was still known, and could be read without difficulty.

[Footnotes]

¹ Madden, Coins of the Jews (ed. 2, 1881), pp. 67 ff., 198 ff., 233 ff.

[page xii -- Introduction ]

In the characters represented hitherto, no tendency to modification in the direction of the modern square type has been observable. Such a tendency first manifests itself in the Aramaic alphabet, and may be traced most distinctly in Aramaic Inscriptions from Egypt. Plate III is a facsimile of the 'Carpentras stele¹,' a monument carved in limestone, the early history of which is not known, but which is now deposited in the Bibliothèque et Musée d'Inguimbert in the town of Carpentras (dép. Vaucluse) in France. The monument is a funereal one: the representation above the Inscription exhibits the embalmed body of the deceased, a lady named Taba, resting on the lion-shaped bier, and attended by the jackal-headed Anubis at the feet, and by the hawk-headed Horus at the head, with the four customary funereal vases beneath. The figures stationed as mourners at a little distance from the head and feet of the bier are Isis and Nephthys. The first three lines of the Inscription are about 9½ inches long; the height of the letters is ⅜ of an inch, or a little more.

The Inscription (=CIS. II. i. 141 = NSI. No. 75), in square characters, is as follows:--

I.e. 1. Blessed be Taba, the daughter of Taḥapi, devoted worshipper of the God Osiris.

2. Aught of evil she did not, and calumny against any man she never uttered.

3. Before Osiris be thou blessed: from Osiris take thou water.

4. Be thou a worshipper (sc. before Osiris), my darling; and among the pious [mayest thou be at peace !].

‭1. ‮תְּמָנְחָא‬; Monḥ is‬ an Egyptian word, meaning perfect, pious; the prefix ta (t’) is the fem. article. זי=Heb. זה: the demonstrative with the force of a relative, as regularly in Aramaic. But זי (= Arab. ذُو) is usually hardened to דִּי in Aram. (Dan. Ezr. passim); the same form,

[Footnotes]

¹ Plate LXIV in the Palaeographical Society's Volume.

[page break]

[Blank page]

[page xiii -- § 1. Egyptian Aramaic Inscriptions ]

however, recurs in Plate V, lines 1, 3, 5, and, as is now known, is the form all but uniformly found in Egyptian Aramaic¹.   2. מִנְדַּעַם something is the oldest extant form³ of the word which appears in Mandaic as מינדאם, in the Targums ‭as ‮מִדַּעַם‬⁴, and‬ in Syriac as ܡܶܕܶܡ: comp. ZDMG. xxxiv. 568, 766. בְּאִיש is the older form of the Syr. ܒܻܝܫ evil: comp. באיש to be evil in the Targums, Gen. 21, 11, and often, בישא (emph.) evil. עֲבַדַת and אֲמַרַת are the usual Aram. forms of 3 fem. pf. כַּרְצֵי must correspond to what is usually written in Aram. as קרצי (see Dan. 3, 8. 6, 25); in Mandaic, however, the root is written כרץ; and comp. Syr. ܟܰܫܳܛܳܐ = Heb. J"|$L>, and Mand. N35J>13 :=jfcC.o!o:=Heb. Blpp. The term will be used here in the derived sense of calumny (though this explanation is not free from objec tion) 5 . nn cannot mean perfect (n?3FI) because adjectives of this form are very rarely derived from verbs y"y (the Aram, form is <+*), and because, as the subj. of mK, we should expect the emphatic nnon. If nnn = Syr. ^br=Heb. OS?, as in Ezr. 5, 17. 6, i. 6. 12, it must mean there, yonder, the speaker being conceived as in the world beyond the grave, and therefore referring to this earthly life as " yonder." This seems, however, rather forced : and it is perhaps better to adopt Lagarde s suggestion that nn = Syr. j> ofcoo (rad. yxaol) " ever" (Dr. Wright). The word must be allowed

This page is released under the CC0 1.0 license, which spells out in as much legal detail as possible my desire to waive any copyright I have over the work. Feel free to use it as you like.