Original Hebrew of a Portion of Ecclesiasticus
A. E. Cowley, Ad. Neubauer
Judaism
1:37 h
This is a scholarly monograph from the late 19th century on one of the Jewish non-canonical Biblical books, Sirach, also known as 'Ecclesiasticus,' not to be confused with the canonical book Ecclesiastes. This particular paper has been cited because it has a section on the 'Alphabet of Ben Sira,' a set of Talmudic Jewish proverbs, each of which begins with successive letters of the Hebrew alphabet. What the general interest reader will find at the core of this otherwise very dry dissertation is an exquisite example of Rabbinical era Jewish wisdom literature.
Original Hebrew of a Portion of Ecclesiasticus
edited by
A. E. Cowley, M.A., and AD. Neubauer, M.A.
[1897]

Every bird dwelleth according to his kind, and (so doth) man according to his like.—
Ben Sira’s Proverbs, XXIV, p. xxii.

This is a scholarly monograph from the late 19th century on one of the Jewish non-canonical Biblical books, Sirach, also known as ‘Ecclesiasticus,’ not to be confused with the canonical book Ecclesiastes. This particular paper has been cited because it has a section on the ‘Alphabet of Ben Sira,’ a set of Talmudic Jewish proverbs, each of which begins with successive letters of the Hebrew alphabet.

What the general interest reader will find at the core of this otherwise very dry dissertation is an exquisite example of Rabbinical era Jewish wisdom literature. What won’t be found here is the Midrash of the Alphabet of Ben Sira, which is in some demand because of a passage about the early Hebrew goddess-figure Lilith. The Midrash of Ben Sira is commentary built on the Alphabet of Ben Sira, and it reputedly includes a number of notoriously transgressive stories. This Midrash has apparently not yet been fully translated into English, and when it is, that translation won’t be public domain.

Production notes: there is much in this document which would be very difficult to convert to text format, due to the limitations of current OCR technology. I have omitted several portions, particularly: the Hebrew and Greek text in the proverbs section, a Latin text of Sirach, a vocabulary, and the Hebrew, Greek and Syriac text from the Sirach section. However, as previously noted, I did transcribe the complete Hebrew text of the Alphabet of Ben Sira.


Preface

IN editing the recently discovered Hebrew fragments of the book of Ben Sira, we have limited our aim to presenting the original text with as little delay as possible, and at the same time giving in a convenient form the materials for further study. A full commentary, as well as a detailed comparison of the versions, must be left for the future. We shall therefore not discuss the author’s full name, or the date of his composition or of the Greek and Syriac translations. For the literature on these points the reader is referred to Schürer’s admirable work on ‘The History of the Jewish People in the time of Jesus Christ .’ In what follows we shall confine ourselves to some remarks on what is known, from Jewish sources, of Ben Sira and his writings.

It is now generally admitted that Jesus, son of Sirach (Σειράχ, ‏סירא‎of Jerusalem, wrote his ethical work (usually quoted as ‘the book of Ben Sira in Hebrew, between 200 and 170 B. C. in Jerusalem. It was translated into Greek by his grandson, as stated in the prologue, from which we also gather that the version was made from the Hebrew, in the year 132 B. C. The Hebrew of the present fragment is classical, not Rabbinical: still less is it an Aramaic dialect, such as that of several of the passages quoted in both Talmuds (the Palestinian as well as the Babylonian), in the Midrashim, and in later Hebrew writings.

In early times the book seems to have hovered on the verge of the canon, or to have been included among the ‏כתובים‎ (Hagiographa, see p. xxii below), since quotations from it are introduced by ‏שנאמר‎ (as it is said), a phrase applied only to the sacred writings. Although afterwards excluded from the canon by the Rabbis, it continued to live and to be appreciated both in the Palestinian and the Babylonian schools, as is proved by the fact that the text was constantly quoted either in the original or in a Rabbinical or Aramaic form. The Rabbis who lived before the destruction of the Temple used it without acknowledgement in the ‘Sentences of the Fathers’ (‏פרקי אבות‎, the earliest production of Rabbinical literature), while others quote from it either expressly under the name of Ben Sira, or anonymously, or else base their maxims upon it. Rabbi Akiba and Ben Azai borrowed from it verbatim, and there is reason to believe that some apocryphal books were influenced by it Thus the official exclusion from the canon did not involve destruction, as in the case of some Christian uncanonical Gospels and Acts: the book of Sirach was allowed to be freely read, but it was regarded merely as literature and not as sacred Passing on to the later history of the book, we find that S. Jerome (fourth century A. D.) possessed a Hebrew copy, although he did not translate it. That the book continued to be known, to individuals at least if not generally, is proved by the passages quoted from it (in a language already debased), by the Rabbis of the fifth and sixth centuries, in the later Midrashim of the seventh and eighth centuries (as the Tanḥuma), and in the sayings collected by R. Nathan in the ninth century Zunz (op. cit., p. 108, end of note e) believes that the early liturgist, R. Eleazar haq-Qalir, borrowed from Sirach (l. 5-8) in his liturgy for the day of Atonement, in praise of the High Priest. Simultaneously some of the sayings of Sirach are quoted by the Babylonian doctors in an Aramaic form

For the tenth century we are on even firmer ground as to the existence of the book in its original language. R. Seadyah (‏סעדיה‎) Gaon, of Bagdad (920 A. D.), and of the Fayyum in Egypt, was blamed by the Qaraites for sending out missives written in Hebrew provided with vowel-points and accents. They reproached him with endeavouring to give to his correspondence an appearance of holiness equal to that of the Biblical text, since the vowel-points and accents were supposed, according to tradition, to have been given with the Law on Mount Sinai. In answer to this accusation Seadyah states that these additions to the text are found also in copies of Ben Sira, in the book of the Wisdom of Eleazar ben Irai (Iri and in the scroll of the Hasmoneans In the course of his defence he quotes seven (or rather eight, see note 2 below) genuine sayings of Sirach in classical Hebrew, so that it may be concluded that the book was at his disposal in the tenth century. The mentions of Ben Sira after this date are scanty and uncertain. R. Nissim ben Jacob (eleventh century) of Kairowân, in Tunisia, makes a quotation, which however he may have derived from Seadyah. In the eleventh century, according to Reifmann signs of Sirach’s influence appear in the collection of sayings entitled ‏מבחר הפנינים‎ (Choice of Pearls), attributed to the famous poet and philosopher Solomon ben Gabirol (Avicebron). The same scholar also finds traces of the influence of Sirach (ii. 18) in the Jewish daily prayer, and (xlix. 10 and 11) in the hymn for the outgoing of the Sabbath. He contends further that Sirach has an allusion (i. 2) to Aristotle and the doctrine of the eternity of matter, and that even Spinoza was perhaps influenced by Sirach (xliv. 34). These suggestions, as well as the inference (from the Aramaic form of the proverb, No. LIV below), that he was an Essene, are, to say the least, not convincing. There is no direct trace of the existence of the Hebrew Sirach in Spain, Provence, or among the Rabbis of France, the Rhine-land, and Germany. Rashi the authors of the glosses on the Talmud (‏תוספות‎), and even Maimonides, did not possess the book; and later Rabbis, who mention sentences from it, most probably quoted second-hand from older authorities. No doubt it might be said, though the supposition is not very probable, that all the quotations from Sirach were made from memory, and that they were derived from oral tradition. Recent discoveries however have removed all uncertainty on this point.

Mrs. Lewis, who brought to light the now famous codex of the Syriac Gospels in the convent on Mount Sinai, some time ago acquired some manuscript fragments in the East among which Mr. S. Schechter, Reader in Talmudic in the University of Cambridge, recognized one leaf as containing a fragment of Sirach (xxxix. 15 to xl. 7) in Hebrew, which he published with English translation, introduction, and notes in the Expositor for July 1896,(p. i seqq.). Through the kindness of the owner we have since been allowed to make a fresh examination of the leaf, and have found reason to alter some of the readings accepted by Mr. Schechter (see the Hebrew text, pp. 2 and 4). Although the leaf is mutilated in places, the parts which are still intact are abundantly sufficient to show the character and style of the composition, and to convince critics that the text is original and not a translation. After pointing this out, Mr. Schechter rightly adds: ‘Its correspondence with the versions changes almost in every line, agreeing in some places with the Greek, in others with the Syriac. In other places, again, it agrees with neither of these versions, omitting whole clauses which are to be found both in the Greek and in the Syriac, or offering new readings which have been either misunderstood or misread by the translators. Certain clauses, again, are to be found in our MS. which are wanting in both versions, or are only reproduced by a very short paraphrase. There cannot, therefore, be even the shadow of a doubt that our text represents nothing else but the original. Even the marginal glosses testify to this fact. Such differences of plena and defectiva as ‏צורך‎ and ‏צרך‎, or such fine variants as ‏פיו‎ and ‏פיהו‎, cannot possibly have been suggested by any translation, and could only have been made from some other copy of the original.’

Almost simultaneously the Bodleian Library acquired, through Professor Sayce, a box of Hebrew and Arabic fragments, among which we recognized another portion of the same text of Sirach, consisting of nine leaves, and forming the continuation of Mrs. Lewis’ leaf, from chapter xl. 9 to xlix. 11. These fragments cannot be part of the copy mentioned by Seadyah, since they are not provided with vowel-points or accents, and also because the writing is not of the tenth century, but of the end of the eleventh at the earliest, as may be seen from the facsimiles. The MS. does not seem to us to have been written by a Qaraite. There are in both fragments marginal notes giving the variants of another copy of Sirach, or more probably of two other copies. These copies were however incomplete, the marginal notes giving their variants only as far as chapter xlv. 8 (see note in loco), and on xlvii. 8 and 9. In the Bodleian fragment there are also at least two Persian glosses (ff. 1 and 5b), which point to its having been written in Bagdad or Persia, possibly transcribed from Seadyah’s copy. The MS. is written on oriental paper, and is arranged in lines, eighteen to the page (in Mrs. Lewis’ leaf one line is cut off), and the lines are divided into hemistichs. There is no indication of chapters, but a line is left blank occasionally, as shown in our printed text. The MS. is unfortunately damaged in many places, which we have marked by clots, showing approximately the number of letters missing, and by[] when letters are supplied. Our object being however to give the text of Sirach as we found it, we have carefully restricted conjecture to its narrowest limits. In some cases we have preferred to leave a lacuna, where either the space in the MS. did not allow of what seemed the obvious word, or some letter such as ‏ל‎, ‏ז‎, or ‏ק‎ was excluded; see e. g. xlv. 13b. In every case a letter about which we felt there could be any reasonable doubt, has been marked with a horizontal stroke, thus ‏א‎̅. On some orthographical peculiarities of the MS. see the note appended to the glossary, p. xxxvi. As regards the translation again, we have deemed it our duty as editors of a unique manuscript, to express the text faithfully, and not to adopt conjectural readings, except where the text yielded absolutely no sense. Usually, indeed, the meaning is clear; but passages occur which, from whatever cause, are obscure, and we cannot feel confident that we have seized the sense of all of them. A(?) in the translation indicates doubt either as to the reading or the rendering. There are sufficient indications that the text is not. everywhere in its original purity, and we do not doubt that (as in many parts of the O. T.) cases will be found in which a purer reading has been preserved by one or other of the early versions; but a detailed comparison of the Hebrew text and the versions, and a discussion of their comparative merits, must, we think, be left to a commentary, as well as to a time when, we may hope, more of the original shall have been recovered. We have noted, lastly, the more important places in which the language is coloured by reminiscences of the Old Testament.

The language, as already observed, is classical Hebrew, the syntax displaying no traces of the peculiar New-Hebrew constructions, such as occur, for instance, so frequently in Ecclesiastes, though the vocabulary has an admixture of late or Aramaic words or expressions, such as might be expected from the date at which the author wrote. The latter, together with other words not occurring in Biblical Hebrew, will be found collected in the glossary (p. xxxi). The style is occasionally a little heavy, but this may sometimes be due to corruption of the text. Otherwise (especially chap. xliv. ff.) it is remarkably easy and flowing. It stands throughout on an altogether higher level than that, for instance, of Chronicles, Ecclesiastes, or the Hebrew parts of Daniel. We know from Ecclesiastes that the New-Hebrew idiom was in process of formation at this time, and it is evident that both New-Hebrew and Aramaic words were current in the Hebrew with which the author was familiar; but the predominant character of his style is nevertheless pure and classical. The marginal readings are often interesting: the variations which they indicate are frequently considerably greater than those noted by the Massorites in the O. T., and resemble rather the various readings often presupposed by the LXX, while at other times they are noticeable as giving an Aramaic equivalent for a Hebrew word in the text. Sirach’s position with regard to the New-Hebrew would no doubt be made clearer by the discovery of the originals of other apocryphal books, such as Judith, Maccabees i, Enoch, and the Psalms of Solomon. Finally the theory that he wrote his proverbs in metre is not supported by the newly-recovered text: the lines are very variable in length, and there is no indication that the author sought to adapt them to a uniform metrical scheme.

In the present edition we give:— (a) The Hebrew text, with the marginal notes and glosses arranged as in the MS.(b) The English translation of the Hebrew, adopting as far as possible the diction of the revised version of the O. T.(c) The Syriac version (which was made from the Hebrew), according to Lagarde’s edition, a blank space indicating that the translator, or copyist, omitted a passage.(d) The Greek translation, according to Dr. Swete’s edition, the blanks again indicating such omissions. The uncertain condition of the Greek text is well illustrated by Hatch, and will strike the reader on even a slight examination. Its value for comparative purposes is further lessened by the translator’s tendency to paraphrase, as is the case also with the Syriac.(e) At the end, the Old Latin, according to Lagarde’s edition of the Codex Amiatinus. For more convenient reference we have in all five texts numbered the chapters and verses as in Dr. Swete’s edition, and indicated the hemistichs by letters of the alphabet in order. The Syriac, Greek, and Latin texts are reproduced exactly as in the editions followed. It did not fall within our plan to give the variants of these versions.(f) A glossary of noticeable words and expressions.(g) A list of proverbs attributed to Sirach in Talmudic and Rabbinical literature, with a translation, arranged in the order of the Greek version. Here again we resolved not to add the various readings, since the Talmudic dialect is not the original language of Sirach, and moreover, all the new Talmudic fragments found within the last two years have not yet been collated. For completeness sake we have added the so-called ‘Alphabets’ of Ben Sira, a late composition— probably of the eleventh century or perhaps even later, but containing some genuine proverbs of Sirach, both in the first and second parts The stories given after each proverb in part i. are mostly indecent, and written in mockery of Jewish literature. We reproduce the first ‏אֹבֹ‎ (MS. second), with a translation: for the second (MS. first), we only refer to the numbers in our list of proverbs with which it agrees, ignoring the rest as alien to Sirach. The Alphabets, though a late and unedifying compilation, survived, whilst Ecclesiasticus was completely neglected. A Persian text of them was lately acquired by the British Museum (MS. Or. 4731), and another copy has just been brought by Mr. E. N. Adler from Persia, probably translated from the Constantinople edition (see below, p. xxix).(h) Some specimens of attempted restorations of the original Hebrew by modern scholars confronted with our text. The comparison will, we think, justify the caution and reservation which must be observed in attempting to restore lost works on the basis of ancient translations. In the present instance, for example, both versions prove to be much freer than was assumed to be the case by those who so used them.

In conclusion, we have great pleasure in acknowledging the help of friends who have enabled us to carry through the work in a short time in spite of difficulties. Mr. J. F. Stenning, of Wadham College, rendered valuable aid in deciphering the difficult parts of the MS., including the Cambridge leaf, and in all doubtful places he concurs in the readings which we have adopted in our text. He also revised the Syriac. Mr. E. N. Bennett, of Hertford College, read the Greek: Rev. F. E. Brightman, Librarian of the Pusey House, read the Latin. Professor D. S. Margoliouth has also shown an interest in the work in various ways. We feel, however, specially grateful to the Regius Professor of Hebrew, Dr. Driver. He revised the translation throughout, besides being entirely responsible for the glossary, with the note appended, and almost every page of the book owes something to the judgement and accuracy which he has been always ready to expend upon it.


A List of Works Used

BACHER (W.), Die Agada der babylonischen Amoräer. Strassburg, 1878.