tion of an alphabet consisting entirely of consonants is highly unnatural and inconceivable, can have but little weight."

2. Jewish tradition. To this belongs the direct assertion of most of the Jewish scholars, that the vowels were published orally by Moses, and were written down and fixed by Ezra and the Great Synagogue. The indirect testimony of the manuscripts of the synagogue is of equal importance. The strong injunction not to point those manuscripts, among a people where all sacred knowledge, and every sacred deed, is a matter of tradition, rests on the belief or the knowledge that it was so formerly. Among the Arabians, the vowel signs were first added to the Koran for the convenience of the reader; but the Jews did not venture to disfigure the divine book by any human addition, and afterwards they referred the reader to the pointed text only to prepare him to read the unpointed."

3. To these must be added the evidence of some passages of the Old Testament, which are intelligible only on the supposition that their author read and wrote without vowels. Perhaps this argu

Schriften, Th. ii. No. 1, Or. Bib. vol. ix. p. 82, 88,) by Eichhorn, (Einleit. vol. i.,) and others.

See a collection of the various opinions, and reference to the literature, in Wolf, Bib. Heb. vol. ii. p. 475, vol. iv. p. 214, sqq., with the arguments, pro and con, in Carpzov, Crit. sac. p. 242, sqq., and Löscher, De Causis Ling. Heb. p. 275, sqq. See the arguments in its favor in Walton, Proleg. iii. 39, and those which oppose it, in Bauer, Crit. sac. p. 128, sqq.

a See Herder, Spirit of Heb. Poetry, vol. i. p. 28.

Some later Jews give us the erroneous view that the Cabalists first introduced the custom of reading the Law without points in order to make it ambiguous. They have been followed, in this opinion, by Buxtorf, (De Antiq. Vocal. p. 35, sqq.,) Carpzov, (Crit. sac. p. 267.) See, on the other side, Bauer, Crit. sac p. 142.

Gen. xix. 37; is explained by e patre (2) Here not only the vowels, but the quiescent seem to have been wanting. Gen. xxxi. 47; 73 is explained as hill of testimony, (,) where the usual pronunciation is wholly overlooked. 1.11;, the threshing-floor, or the province of Egypt, is explained by ", mourning of Egypt. The first is, doubtless, to be preferred. Judg. xv. 18; the name derived from the circumstance that Samson cast the jaw-bone out of his hand, and therefore from, to cast. So it would mean jaw-bone-cast. But this could be true only if pointed, —,—and presupposes a derivation from . 2 Kings xxii. 9; it now stands, and there came

(height of the jaw-bone) is

ment may be weakened by the fact that the violent etymologies of the Old Testament often turn on the consonants as well as the vowels; but it is confirmed,

4. By the certain reference which was made in the following age to this earlier period.

II. It may be true that such writings would be difficult to read, and ambiguous; but this fact is no objection to it. A knowledge of the language, especially the habit of using it as a vernacular tongue, compensates for much, and is quite indispensable in reading the modern languages, which are written according to their etymology, (such as the French and English.) But is not the reading of the Talmud very difficult for us? In general, casiness of reading must not be sought among the Orientals. At this day, few nations learn to read more slowly and painfully than the Arabians, and few writings are so little read, or so often misunderstood, as theirs. Even their own learned men will seldom read an unpointed manuscript which they are unacquainted with, without preparation." The frequent mistakes made in reading the Koran mainly gave occasion to the introduction of the vowels."

с

III. In order to lighten this difficulty in some measure, men have resorted to the conjecture that the ancient Hebrews perhaps used the matres lectionis more frequently than they are used in the present text, and that, after the introduction of vowels, these letters again disappeared. And it is true these letters occur more frequently in the unpointed text of the Talmud, Targums, and Samaritans. Without stopping to examine some uncritical arguments, an appeal has some

d

Shaphan the scribe. If the writer in Chronicles had read it with these vowels, he

וַיָּבֹא שָׁפָן אֶת הַסֵּפֶר,so as to read, את would scarcely have interpolated an

and Shaphan brought the book, as now it does, 2 Ch. xxxiv. 16.

a Elias Nisibensis (apud Abraham Echellensis, ad Ebed Jesu, Catal. Libr. Chald. p. 178) says, Non possunt recte legere nisi tanquam divinantes, aut ex traditione. See Ch. Th. Tychsen, 1. c. p. 260. Volney, Simplification des Langues Orient. p. 20.

b Abraham Echellensis, 1. c. p. 227. Jo. Morinus, 1. c. p. 529.

The three letters, 1, .

d See Cappellus, Arcan. Punct. i. 18, 19. Jo. Morinus, 1. c. Exercitat. Bib. xviii. c. 3. Bellermann, Handbuch der Bib. Lit. vol. i. p. 88, sqq., and others, in Löscher, 1. c. p. 297, sqq. Morinus, De Ling. Primæva, p. 346, sqq., (comp. Vitringa, Obs. Sac. p. 73,) and the refutation in Dupuy, Sur les Voyelles de la Langue Heb., in the Memoires de l'Acad. vol. xxxvi. p. 239. Michaelis, Vermischte Schriften, vol. ii. No. 1, § 15-22. Jahn, Einleit. vol. i. p. 337. Bauer, Crit. sac. p. 146.

times been made to the matres lectionis that occur, here and there, in the text, and it is maintained they are the relics of the old orthography. But they occur only in some particular manuscripts, from which they have found their way into the received text: they prove nothing. Some of them are more modern than the vowels, but they are all the production of a fluctuating orthography." The later copyists indulged themselves in these variations to a great extent, as a single glance into Kennicott's collection of various readings will show. But the whole assertion, however it may be limited, is contrary to all the analogy of the old Shemitish writing. The oldest Phoenician inscriptions and coins are uncommonly sparing of these letters, and omit them in cases where the omission very rarely occurs in the Hebrew. The Jewish coins sometimes have the full, sometimes the defective, reading. The same remark is true of the oldest Hebrew writings. The scriptio plena first became prevalent in the period when the language was half dead. It was used merely to render reading more easy. The orthography of the Samaritans, the Talmud, the modern Hebrew, and Chaldee, agrees with this statement. This very distinct gradation in the masoretic text is certainly genuine, for it is founded on the circumstances of the case, and supported by analogy. How could the men who affixed the points at a later age permit so many letters to be removed from the text as that hypothesis would make it appear? And, furthermore, how could we explain the perpetual diversity between the Septuagint and the Hebrew words?

$ 3.

THE VOWELS IN THE SEPTUAGINT, JOSEPHUS, AND OTHERS.

1. We have in the Seventy a valuable evidence of the condition of the new writing at that period when the old Hebrew was completely extinct as a living language. After a careful examination, it appears to me this version proceeded from a text entirely destitute of vowels; and the statement in the previous section is thereby confirmed, for it is not probable that there were vowel characters which were yet not used or known."

See the Excursus on the Phoenician Language, No. 3.

See collections of these variations, from all the biblical books, in Morinus, De Ling. Primæva, p. 385-396. Cappellus, Crit. sac., ed. Vogel-Scharfenberg, p.

It has been objected, notwithstanding, that this version agrees so often and so strikingly with the present pointed text, that the agreement cannot be explained from the context, without we admit there were certain vowel characters in the text; and in particular that the words which occur but once in the Bible, and are distinguished from other well-known words solely by the vowels, are written in this version with perfect correctness, and this could not have been done if the writing character had not given them a hint." But after examining the explanation of these words. in the Seventy, it becomes clear that the context and tradition were their only guides, and these not rarely forsook them."

500-545. Comp. Wepler, Philol. Krit. Fragmente; Cassel, 1783, vol. ii. p. 10, sqq.

a Buxtorf, De Punct. Orig. p. 116, sqq. Pococke, Com. in Hos. Bertholdt, p. 176.

To show the strength or weakness of this argument, the following collection of explanations, given by the Seventy, chiefly relating to the letter N, may be examined: :

1. They continually distinguish between the following words: and N;

.Isa. xiii ,אחים and אחים ;13 .Deut. xxiii , אֲזַנִים and אֹזֶן ; אֲבָל, אֵבֶל, אבל

,locust ,אַרְבֶּה ;2 .Cant. vii ,אָמָן and אָמֵן ;30 .Prov. viii ,אָמוּן and אֱמוּן

21;, where? and, alas! island;, a measure, and, where?

and, net-work;, ashes, and "N, 1 Kings xv. 38, 42;

N, caravan, and, portion; MN, offering, and , &c. But in all these the context makes the distinction clear.

cases,

2. They also make a proper distinction between words of similar sound; e. g. П, aðéos, and П, tozatos; i, belly, and is, magician; DN, also, but, and , nose; where the connection decides the meaning.

3. But where the context is uncertain, or the translator ignorant and careless, they confound words which ought to be distinguished.

Isa. xvii. 11, 9, 16.),

as if it were

80 ;ארל were

logus, as if

ND, ös narię àvdgóлov, for "; (comp. Jer. xvii. ión, Ps. xlv. 10, but ozývaι, in Num. xxiv. 6, and Prov. vii. 17, ;, commonly lapos, but, in Lam. i. 6, xolos, as if it

, commonly λapos, but in plural, Cant. i. 2, N, it is ;, Gen. xlix. 21, orilexos, for ;,, as if

, Ps. vii. 12; ", devoo, for ", Cant. iv. 8; 1, uas, for, Ps. lxxiv. 9. They constantly distinguish, oath, from, these, and b, God; but not from, because both are names of trees, which the con

Besides the context, which, in many places, would decide for them, the traditionary and common explanation would assist them. We need only recur to the versions of the Samaritan codex, which certainly had no points. There are many passages which, at the first glance, render it highly probable there were marks in the text to indicate the pronunciation; but the diversity of the other versions in this respect opposes the supposition. . . . .

2. We come to the same result by observing the method after which the numerous proper names in the Hebrew are translated into the Greek. Sometimes they are pronounced with vowels very different from the original; sometimes according to a different system of vocalization, which, departing from the masoretic custom, is sometimes analogous to that of the Arabic or Syriac." . . .

3. The examples in the fragments of the other Greek versions, and in Josephus, are of the same character. The latter, throughout, betrays an unpointed text, for, wherever he does not use the Seventy, but the Hebrew text itself, he differs from the orthography of the former.

......

4. But if an inquirer is not convinced by what has hitherto been said, and thinks, with Schultens, Michaelis, Eichhorn, and others, that in this period there were certain signs in the text to guide the reader, he would naturally recur to a diacritical point. His opinion will be favored by the Samaritan Pentateuch, which proceeded from a Jewish copy before the time of the Seventy, and is now, at least, furnished with this sign, (though it cannot be proved that it had it so early,) and by the old Syriac character also, which certainly is a descendant of the square letter. The matres lectionis in the Samaritan codex give the same result with those of the Seventy.

$4.

TRACES OF THEM IN THE TARGUMS AND TALMUDS.

1. The agreement between the oldest Targums and the vowels of the present text is very striking. We should be compelled to conjecture that the Palestine scholars, at least, had a pointed text at that time, if

text does not separate. See the proof-texts in Trommii, Lex. Heb., at end of his Concord. e LXX., Int.

a See Hilleri Onomast. sac. p. 706, sqq. Masch, Bibliotheca sac. pt. ii. vol. ii. p. 35, sqq.