§ 5. OTHER ANCIENT VERSIONS

1. The Syriac Version.

It is impossible to fix the date of the Syriac version of our book with any certainty; the earliest known MS. (Cod. Mus. Brit. 12142) belongs to the sixth century, but this MS. contains already a very large number of scribal errors, which points to a long previous history; it seems, however, to be the parent of all other extant Syriac MSS. of Sirach, for its corruptions occur in all of them. Wright, in speaking of the Syriac translations of the Old Testament Apocrypha, the dates of which are quite unknown, says that ‘it seems tolerably certain that alterations were made from time to time with a view to harmonizing the Syriac text with that of the Septuagint’,2 a process which Burkitt thinks ‘may have begun as early as the episcopate of Palut (about a.d. 200)’, which would imply the existence of a Syriac version some time previous to this date.

Although some scholars long ago sought to show that the Syriac version of Sirach was a translation from the Hebrew, their contention was combated by Syriac scholars, who maintained that it was translated from the Greek.3 The discovery of the Hebrew text has, however, definitely settled the matter; if there was reason to believe, as was certainly the case, that the Syriac text itself presented indications of its having been translated from Hebrew and not from Greek, there is absolutely no doubt about this now that we can compare the Syriac with the Hebrew. Nevertheless, the Syriac translation was not made from the original form of the Hebrew, though from a form which seems to have been in many respects nearer to the original form than that represented in the recently found Hebrew MSS. This fact makes the Syriac version valuable for correcting, where necessary, the Hebrew text in the form in which we now have it; and for those large portions of the book of which the Hebrew text has not been found the Syriac is, of course, indispensable. Another fact which makes the Syriac version valuable is that it contains a number of verses and parts of verses which are only found elsewhere either in the Hebrew alone, or in isolated Greek MSS., in some few cases also in the Old Latin version.4 ‘In some instances the Syriac has retained the correct text where both the Hebrew and the Greek agree in having gone astray. But in a considerable number of passages the Syriac is not a translation of the Hebrew, but of the Greek;5 it is possible that the reason of this was that in such cases the Greek version represented what the original Syriac translator believed to be the reflection of a more original form of the Hebrew than that which he had before him; or else, and this is more probable, it may be that the Syriac, as we now have it, has been corrected on the basis of the Greek; this would have been a very natural proceeding (even if a comparatively speaking pure Hebrew text had been available) at a time when the Greek Bible was regarded in the Christian Church as more authoritative than the Hebrew. That the Syriac translator of Sirach was a Christian seems more than probable. The Greek MS. or MSS. which the Syriac translator made use of contained elements representing the secondary Greek text, and it was a text which had undergone deterioration in other respects.’6 In any case, the Syriac version is one which has a distinct value; nevertheless it must be used with caution, for, in spite of what has been said about its usefulness and importance, it has some grave blemishes which must be taken into consideration when utilizing it. Smend says it is the worst piece of translation in the whole Syriac Bible, though in many cases it is uncertain in what proportion its mistakes are due to the translator himself, or to the Hebrew text which he had before him, or to some deteriorated Greek text which he utilized, or to textual corruptions which crept in during the process of transmission. But, however this may be, the fact remains that the work of translation has been done carelessly and without much trouble having been expended upon it; paraphrases abound; sometimes they are of a purely arbitrary character, at other times they apparently represent what the translator believed to be the general meaning of the original, which he did not understand in all its details; in yet other cases these paraphrases were evidently due to the desire to give a Christian sense to a passage. But perhaps the most serious blemish in this version is the large number of omissions; Smend says that these amount to 370 stichoi, or one-ninth of the whole book. In many cases it is evident that the Syriac translator had what seemed to him good reasons for omitting certain passages; thus, as a Christian he felt justified in omitting such words as these:

Thanksgiving perisheth from the dead as from one that is not,

(But) he that liveth and is in health praiseth the Lord (17:28).

It was probably owing to an anti-Jewish tendency that he omitted 37:25:

The life of a man (numbers) days but few,

But the life of Jeshurun days innumerable.

A similar reason would account for the omission of 38:11, 45:8–14, parts of 50:18–21, and the litany after 51:12, though this last is also omitted in the Greek version. Quite comprehensible are the omissions of 33:26 (G 30:35) and 36:21, 23 (G 26, 28); but why such passages, e.g., as 41:14–42:2, and most of 43:11–33 should have been passed over it is impossible to say, excepting on the supposition that they are difficult ones to translate, and the Syriac translator did not feel inclined to undertake the task.

It will thus be seen that while the syriac version has a distinct value of its own and can certainly not be neglected, it must nevertheless be used with great caution; indeed, the student will be wise never to utilize it without at the same time referring to the Greek. It should be added that in this version the right order of the chapters is preserved.

2. The Old Latin Version.

This is the oldest1 and most important of the daughter-versions of the Greek. Like the Syriac version, while in some respects it is valuable for correcting the Greek, in other respects it presents grave drawbacks. As we shall see later on (§ 8), Jerome left the Latin text of Sirach as he found it—a matter for congratulation, since as the version now stands it contains many really ancient elements which would probably have been lost altogether if Jerome had undertaken a translation of his own. That it contains, as we have already seen (p. 281), the chapters in the right order is also a fact of importance. But the text of the Old Latin version has come down to us in a deplorable condition, added to which it has the further disadvantage of having been made from a Greek text which was in a worse condition than that represented by any extant Greek MS. Moreover, the Old Latin text is full of scribal errors, and many arbitrary alterations have been introduced; quotations from this version in the writings of the Latin Fathers are of little use for emending its text. Emendation is made the more difficult in that the original translation was apparently subjected to constant correction on the basis of different Greek texts; one example out of a great many may be given: in 13:8 the Greek text runs:

πρόσεχε μὴ ἀποπλανηθῇς (V 106 248 253 Syro-Hex add τῇ διανοίᾳ σου),

καὶ μὴ ταπεινωθῇς ἐν εὐφροσύνῃ (248, &c. add καρδίας) σου.

For this the Latin has:

Attende ne seductus

In stultitiam humilieris.

Noli esse humilis in sapientia tua,

Ne humiliatus in stultitiam seducaris (= vv. 10, 11 in Latin).

In cases like this the question arises as to whether the additions have been inserted from other Latin texts, or whether they are doublets due to the incorporation of marginal notes into the text; in other words, do they represent different Greek texts from which Latin translations were made, or are they merely Latin variations of one and the same Greek text? It is by no means always possible to decide which, a fact which materially increases one’s difficulties when utilizing the Latin version.

The question as to whether the Old Latin version was made from a MS. representing the primary or secondary Greek text is one of extreme complexity; at first sight one would feel impelled to postulate the secondary Greek text as the basis of the Latin version, but the fact that many of the additions belonging to the secondary Greek text are preserved in the Syro-Hexaplar but not in the Old Latin goes to show that the latter cannot have been made directly from the secondary Greek text. And yet, as Smend has shown, the copy from which the Old Latin was made was more influenced by the secondary Greek text than any other known Greek MS., though that copy did not in itself represent the secondary Greek text; for, as Smend says, ‘trotz aller sekundären Elemente, die die Vorlage enthielt, und trotz aller Bearbeitung, von der der ursprüngliche Text des Lateiners betroffen sein mag,—durch Massenbeobachtung lässt sich nachweisen, dass die Vorlage des Lateiners im Wesentlichen der griechische Vulgärtext war, den der Lateiner nicht nur in alten guten Lesarten, die freilich auch aus Gr. II’ (i.e. the secondary Greek text) ‘stammen könnten, sondern auch in höchst sekundärer Entartung vor sich hatte’ (op. cit., p. 124). The proof of this is minutely worked out by him. The conclusion which Smend draws from this complicated state of affairs is that the foundation of the text from which the Old Latin Version was made was the primary Greek one, but that in that text was incorporated a later recension of the secondary Greek text, the offspring, perhaps, of a Hebrew original.1 And it must be confessed that only on this hypothesis can all the phenomena of the Old Latin version be explained.

3. The Syro-Hexaplar.

This is the Syriac version made by Paul of Tella from the Greek (c. a.d. 616). ‘If we retain the designation Syro-Hexaplar,’ says Nestle, ‘we must bear in mind that Sirach had no place in Origen’s Hexapla; but in one particular respect this. Syriac version reminds us of the Hexapla; one of the critical marks of Origen, the asteriscus, appears also in Sirach, at least in its first part up to chap. 13.’2 Hart, on the other hand, remarks: ‘Origen valued the Book of Jesus Ben-Sira, and its text required a settlement. It seems reasonable to accept the evidence—direct and indirect—as it stands, and to conclude that he attempted to purge the current Greek version of its accretions, and that his disciples removed them bodily, and sometimes parts of the true text with them’ (op. cit., p. 359). It is true we are nowhere told that Origen incorporated the books of the Apocrypha in his Hexapla, but the way in which he quotes from them, speaking of them as ‘Holy Scripture’ (see below, § 8), would naturally lead to the supposition that he did so incorporate them. There is also the fact that in the Syro-Hexaplar the Book of Baruch undergoes much the same treatment with regard to the Hexaplaric signs as the canonical books. As Smend says: ‘The excellence of Syro-Hexaplaric Sirach text would not be unworthy of Origen.’ In general the text of this version follows very closely a MS. with which Cod. 253 was intimately related, but the translator also utilized the Peshiṭta when for one reason or another he found it convenient to do so. The translation is in parts very free. There are a number of marginal notes which were presumably taken from the Greek MS. which the translator had before him; these not infrequently present the better reading.

4. The Sahidic Version.

The MS. containing this version, which is based on a Greek text closely related to the Greek uncials, has suffered a good deal of mutilation; the Prologue is almost entirely wanting through this cause; in addition to this a good many stichoi are omitted. Other Coptic versions of more or less value are the Bohairic and Akmimic; only fragmentary remnants of these are extant. For the published texts of them see Smend, op. cit., p. 130 f.

5. The Ethiopic Version.

This version is rendered from the Greek, of which it is often a literal translation, but in his desire to make the meaning of the original before him clear the translator often interprets, i.e. he gives a paraphrastic rendering. Smend (op. cit., p. 132) gives as an example of this 24:21, where for the Greek:

οἱ ἐσθίοντές με ἔτι πεινάσουσιν, καὶ οἱ πίνοντές με ἔτι διψήσουσιν,

the Ethiopic has:

They that eat me, eat me and are not satisfied,

And they that drink me, drink me and are not satisfied.

The value of this version is not infrequently marred by the fact that its renderings only partially represent the Greek, the reason being presumably that the Ethiopic translator did not really understand what was before him and made a guess at the meaning. Smend quotes Dillmann (who has edited this version, V.T. Aethiopici, tom. v, Berol. 1894) to the effect that the Ethiopic has been subjected to revision in later times on the basis of the Greek text; he does not believe it possible to restore the Ethiopic text to its original form from the MSS. which are now extant. Upon the whole the Ethiopic follows the text of Cod. B; the MS. from which the version was made, however, represented in some cases a purer, in others a more corrupt, form of text than that of Cod. B. Here and there the Ethiopic is of value for correcting the Greek.

6. The Armenian Version.

This version exists in two forms, one translated from the Old Latin, and the other from the Greek; the former is but of small value for text-critical purposes. The latter is of more use, but a good deal of the text is wanting, viz. 36:1–38:14, 43–51, besides a number of isolated passages. On the other hand, it has some additions which are singular to it; see Herkenne, Armenischer Sirach, p. 30 ff.

7. The Slavonic Version.

One point of importance regarding this version is that it sometimes agrees with the Old Latin against all extant Greek authorities; moreover, in agreement with the Old Latin and against all known Greek MSS. (with the exception of Cod. 248) it has 30:25–33:13 a in the right place. According to Margoliouth it ‘follows a text similar to that of the Complutensian edition, but with only a portion of the additions’.1 It has been revised from the Syro-Hexaplar.

8. The Arabic Version.

This is a translation of the Peshiṭta (Syriac Vulgate). ‘The translator’, says Smend, ‘was not concerned to offer a careful translation of his copy; he desired rather to present an elegantly-written Arabic book.’ His translation is, therefore, throughout a free one, and he inserts additions of his own. The text, or that of the MS. from which it was made, has been influenced by the Greek. But the manifold lacunae of the Syriac version recur in the Arabic. The version is of but small value.