OUR BIBLE & THE ANCIENT MANUSCRIPTS by SIR FREDERIC KENYON - formerly Director of the British Museum - © Sir F Kenyon 1895. First published Eyre & Spottiswoode 1895. - fourth edition 1939. - Prepared for katapi by Paul Ingram 2003.


HOME| Contents | << | The Samaritan Pentateuch | The Septuagint & Other Greek Versions | Other Eastern Versions | The Latin Versions | Condition of the OT Text | (LXX) >> |

IN August, 1883, the world was startled by the announcement of a discovery which, if it were authentic, seemed to go far towards bridging the great gap in our knowledge of which we spoke at the end of the last chapter. This was no less than some fragments of a manuscript of the Old Testament purporting to have been written about eight hundred years before Christ, which their owner, a Jew of the name of Shapira, stated that he had obtained from some Arabs about five years before. The material was old leather, and the writing was similar to that of the Moabite Stone. The contents were striking enough. They purported to be portions of the Book of Deuteronomy, but with many remarkable variations. To the Ten Commandments was added an eleventh, and the language of the others was altered and amplified. In these strips of leather there was enough to cast doubt upon the whole of the received text of the Old Testament and to discredit the whole science of textual criticism. The sensation, however, lasted only a few days. Evidences of forgery soon began to pour in; and the final blow was given when it was shown that the strips of leather on which the characters were written had been cut from the margins of an ordinary synagogue roll.

There is, indeed, no probability that we shall ever find manuscripts of the Hebrew text going back to a period before the formation of the text that we know as Massoretic. We can only arrive at an idea of it by a study of the earliest translations made from it; and our task in the present chapter is to describe these translations in turn.

 1. The Samaritan Pentateuch.

Samaritan Pentateuch

Origin | Discovery | Character | Manuscripts

Its Origin.

The version of the Old Testament, which possesses the longest pedigree, is that which owes its existence to the Samaritans. Strictly speaking, it is not a version at all, as it is in the Hebrew tongue, though written in a different character from that of the extant Hebrew MSS.

It is written in the old Hebrew character, such as it was before the adoption by the Jews of the square characters, as described in the last chapter (p.40). The precise origin of this separate Samaritan Bible has been a subject of dispute; but the most probable account is that it takes its rise in the events described in Neh.i.23-30 - namely, the expulsion by Nehemiah of those Jews who had contracted marriages with the heathen. Among those expelled was a grandson of the high-priest Eliashib, whose name, as we learn from Josephus, was Manasseh. This Manasseh, in indignation at his expulsion, took refuge among the Samaritans, and set up among them a rival worship to that at Jerusalem. The Samaritans, whom we know from 2 Kings xvii.24-41 to have been foreigners imported into the country of the Ten Tribes by the king of Assyria, and there, presumably, to have mingled with the scanty remnant of Israelites, had at first incorporated the worship of Jehovah, as the God of the land, into the worship of their own gods; and later, on the return of the Jews from captivity, had been willing to join in the rebuilding of the Temple at Jerusalem, but had been refused permission. Since this repulse they had been bitterly hostile to the Jews, and the schism of Manasseh gave them a head and a rival worship, which embittered and perpetuated the quarrel. Manasseh obtained leave from Darius Nothus, king of Persia, to set up a temple on Mount Gerizim, which became the centre of the new religion and the rival of Jerusalem. He had brought with him, it is believed, the Hebrew Pentateuch, and this, with certain alterations (notably the substitution of Gerizim for Ebal in Deut.xxvii.4 as the hill on which the memorial altar should be placed), became the sacred book of the Samaritans. As we have seen in the last chapter, probably this was the only part of the Old Testament that had at that time been definitely recognised as inspired Scripture by the Jews themselves; and when the Prophets and Hagiographa were subsequently added to the Canon, the Samaritans refused to accept them. They refused also to accept the square Hebrew characters adopted by the Jews; and we may be quite certain that they would pay little respect to any alterations in the text, if such there were, which were made by Jewish scribes and scholars after the date of the original secession.

Its Discovery.

So far, then, it appears as if we had, in the Samaritan Pentateuch, an invaluable means of testing the extent of the variation which the Hebrew text has undergone since the days of Nehemiah. We have an independent tradition, coming down from about BC 408 (the date of Manasseh's secession), without any contact with the Hebrew text, preserving the original form of writing, and thereby avoiding one considerable source of possible error and corruption. No wonder that when, in 1616, the first copy of the Samaritan Bible came to light many scholars thought that they had obtained evidence for the original text of the Old Testament far preferable to that of the Hebrew manuscripts. The Samaritan community had existed from the days of its first settlement by Sargon of Assyria until then, and it exists still, a little community now of less than a hundred persons, settled at Nablus, the ancient Shechem, still observing the Mosaic Law, and still celebrating the Passover on Mount Gerizim; but none of their sacred books had come to light until, in that year, a copy was obtained by Pietro della Valle. Several other copies have since been secured by travellers and are now in European libraries. The first printed edition was issued in the Paris Polyglot Bible in 1632, and for generations a hot controversy raged among Biblical scholars as to the comparative value of the Samaritan and Hebrew texts. At length, in 1815, it was settled, for the time, by an elaborate examination of all the variations by the great Hebrew scholar Gesenius, whose verdict was wholly against the Samaritan version. He divided the variations into groups, according to their character, and argued that in hardly a single instance was a Samaritan reading to be preferred to that of the Hebrew. This opinion has held the field until recently; but there seems to be a disposition now to question its justice.

Its Character.

The Samaritan version has been estimated to differ from the Hebrew in about 6,000 places. The great majority of these are of very trifling importance, consisting of grammatical alterations or the substitution of Samaritan idioms for Hebrew. Others (as in Deut.xxvii.4, quoted above) are alterations of substance, so as to suit Samaritan ideas of ritual or religion. Others contain supplements of apparent deficiencies by the help of similar passages in other books, repetitions of speeches and the like from parallel passages, the removal of obscurities or insertion of explanatory words or sentences, or distinct differences of reading. In all these latter cases there may evidently be two opinions as to whether the Samaritan or the Hebrew reading is preferable. The apparent deficiencies in the Hebrew may be real, the obscurities may be due to error, and the Samaritan text may be nearer to the original language. This probability is greatly increased when we find that in many passages where the Samaritan version differs from the Hebrew, the Greek Septuagint version (of which we shall speak presently) agrees with the former. For example, the Samaritan and Hebrew texts differ very frequently as to the ages of the patriarchs mentioned in the early chapters of Genesis. Gesenius classified these variations as alterations introduced on grounds of suitability; but it is at least possible that they are not alterations at all, but the original text, and that the numbers have become corrupt in the Hebrew text; and this possibility is turned into a probability when we find the Septuagint supporting the Samaritan readings.
There is no satisfactory proof of either the Septuagint or the Samaritan text having been corrected from the other, nor is it in itself likely; and their independent evidence is extremely difficult to explain away. Hence scholars are now becoming more disposed to think favourably of the Samaritan readings. Many of them may be errors, many more may be unimportant, but there remain several, which are of real value. The editors of the Variorum Bible give thirty-five variations of the Samaritan text in the five books of the Pentateuch as being either equal or superior to the Hebrew readings. Among these may be mentioned, for the sake of example, Gen.iv.8, where the Samaritan has "Cain said to Abel his brother, Let us go into the field"; xlvii.21, "As for the people he made slaves of them," instead of "he removed them to cities"; Exod..40, the 430 years of the sojourning of the children of Israel are said to have been in Egypt and in Canaan (thus agreeing with Gal.iii.17), instead of in Egypt only; Num.iv.14, the following words are added at the end of the verse, "And they shall take a cloth of purple, and cover the laver and his foot, and put it into a covering of seals' skins, and shall put them upon a frame"; and in Deut.xx.35 the first half of the verse runs "against the day of vengeance and recompence; against the time when their foot shall slip." These are perhaps the most notable of the Samaritan variants, and it is observable that in every case the Septuagint confirms them. The general result of the comparison of this and the other versions with the Hebrew text must be reserved to the end of the chapter; meanwhile it will be sufficient to observe that these variations, though sufficient to arouse our interest, are not serious enough to cause any disquietude as to the substantial integrity of the text of our Old Testament.

Its Manuscripts.

No manuscript of the Samaritan Bible (so far as is known) is older than the tenth century. It is true that the Samaritan community at Nablus cherishes a precious roll, which it maintains to have been written by Abisha, the great-grandson of Moses, in the thirteenth year after the conquest of Canaan; but this story, which rests on the authority of an inscription said to be found in the MS. itself, may very safely be dismissed.

[There is much mystery about this MS. No competent authority has ever examined it, nor have the columns containing the inscription been photographed. An ancient roll used to be shown to visitors, but it was said that this was not the real roll of Abisha, which was kept secret. In 1926, however, all antiquities in Palestine had to be registered with the Department of Antiquities, to secure ownership. At that time I happened to visit Nablus with the Director of Antiquities, and was shown what purported to be the original roll (as well as others), and even the inscription was pointed out to me. This is not written at the beginning or end of the MS., but (according to a special Samaritan practice) was inserted, acrostic-wise, in the middle of several consecutive columns of the text, by isolating selected letters about the middle of each line. Such an inscription could not be inserted later, though it might be possible to alter letters. The roll was of thin vellum (not leather, as one would expect in the case of the age suggested), rather tattered, and had a distinctly medieval appearance to my eye. Subsequently some photographs were sent to me, which were certified by Dr. L. A. Mayer, of the Jerusalem Department of Antiquities, as being taken from the oldest MS. One of these is reproduced as Plate V. Unfortunately they did not include the columns with the inscription. In the expert opinion of Sir A. Cowley, the hand shown in these photographs is of a thirteenth-century type, certainly not materially older. The knobs on the rollers are not relevant, since they are removed when the roll is put away, and can be attached to other rolls. The photographs have been presented to the Department of Oriental Printed Books and Manuscripts in the British Museum. According to Sir A. Cowley, the Samaritan historian Abulfath says that the roll of Abisha was "discovered" in 1355 by the High Priest Phinehas b. Joseph. Probably its real date is not much earlier than that.]

All the existing manuscripts of the Samaritan version are written on either vellum or paper, in the shape of books, not rolls, with the exception of three rolls at Nablus, without any vowel-points or accents, but with punctuation to divide words and sentences. The whole of the Pentateuch is divided into 964 paragraphs.

2. Septuagint & other Greek Versions

3. Other Eastern Versions.

Syriac | Coptic | Ethiopic 

The Syriac Version.

The two versions of which we have hitherto spoken, the Samaritan and the Greek, were made before the institution of Christianity. It is otherwise with all the remaining versions of the Old Testament.  Outside the Jewish and Samaritan communities there was no desire to know the Hebrew Scriptures until Christianity came, preaching the fulfilment of those Scriptures and the extension of their promises to all nations. As the Christian missionaries spread abroad from Judaea into the surrounding countries, fulfilling their Master's last command to preach the Gospel to every people, they necessarily referred much to the history of the nation among which He wrought His ministry, and to the prophets who had prepared His way before Him. Hence there arose a demand for translations of the Hebrew Scriptures into the languages of every country in which Christianity was preached; and the versions of which we have now to speak were all the offspring of that demand. The first of these in geographical nearness to Judaea was the Syriac. Syriac is the language of Syria and Mesopotamia, which lie north and north-east of Palestine, and, with some slight differences of dialect, it was the actual language commonly spoken in Palestine (and there known as Aramaic) at the time of our Lord's life on earth. In the case of the New Testament, as we shall see, several translations into Syriac were made; but of the Old Testament there was (apart from the version of Origen's Hexaplar text, mentioned above, p.59, and some other late translations from the Septuagint, of which only fragments remain) only one, and that the one which is and always has been the standard version of all the Syriac Churches. It is known as the Peshitta, or "Simple" version, but the exact explanation of the name is unknown. It was probably made in the second or third century after Christ; certainly not later, since in the fourth century we find it quoted and referred to as an authority of long standing. A considerable number of copies of it are known, most of them forming part of a splendid collection of Syriac manuscripts which were secured for the British Museum in 1842 from the monastery of St. Mary Deipara, situated in the Nitrian desert in Egypt. Among these is a manuscript dated in the year ad 464, which has the distinction of being the oldest copy of the Bible in any language of which the exact date is known. We thus have direct evidence of the text of this version in the fifth century, and in the century before that we find copious quotations from it in the writings of two Syrian Fathers, Ephraem and Aphraates. 

The Peshitta version originally omitted the books of the Apocrypha, and hence was evidently taken from Hebrew MSS. after the Canon of the Hebrew Scriptures had been finally fixed. It also was originally without the Chronicles, which were added to it (from a Jewish Targum) at a later time. The cause of the omission is not known, and it may have been due simply to a belief that the Jewish history was sufficiently represented by the books of Kings. The whole translation is from the Hebrew, but the translators have been rather free in their renderings, and seem also to have been acquainted with the Septuagint. The books of the Apocrypha (except 1 Esdras and perhaps Tobit) were added at an early date, and they now appear in all the earlier Syriac MSS. which make any pretence to contain a complete Old Testament. The Syriac version of these books is often useful in correcting errors which have found their way into the Greek text. [Especially in the book of Ecclesiasticus, in which the Syriac version must have been made from the Hebrew original; see the Variorum Apocrypha and the editor's preface. On the Hebrew original of this book, see below, p.93.]
At a later date the whole version was revised by comparison with the Septuagint; and hence it is not very trustworthy as evidence for the Hebrew text, and its agreements with the Septuagint cannot be taken with any certainty as independent confirmations of its reading.

The Coptic Versions

Sahidic Codex of Acts (4th cent)(see Plate XXIII). - Coptic is the language which was used by the natives of Egypt at the time when the Bible was first translated for their use.

It is, indeed, a modified form of the language which had been spoken in the country from time immemorial; but about the end of the first century after Christ it began, owing to the influence of the great number of Greeks settled in Egypt, to be written in Greek characters, with six additional letters, and with a considerable admixture of Greek words.  It is to this form of the language that the name of Coptic was given, and it continues to the present day to be used in the services of the Christian Church in Egypt. There were, however, differences in the dialects spoken in different parts ' of the country, and consequently more than one translation of the Scriptures was required. The number of these dialects is still a matter of uncertainty, for the papyri discovered in Egypt of late years have been, and still are, adding considerably to our knowledge of them; but it appears that four or five different versions of the New Testament have been identified, and three of the Old. Two of these stand out as of real importance, the others being mere fragments. 

The Coptic versions of the Bible are more important for the New Testament than for the Old, and it will consequently be convenient to treat of them at greater length in the chapter dealing with the versions of the New Testament. In the Old Testament they were made from the Septuagint, and consequently their evidence is mainly valuable for the purpose of restoring the Greek text, and only indirectly for the Hebrew text which lies behind the Greek. For the student of the Septuagint, however, they should be of considerable service. As it is probable that they were taken from the edition of the Septuagint current in Egypt, which was that of Hesychius, they should give valuable assistance in identifying and recovering the text of that edition. The two most important of the Coptic versions are -

  1. the Memphitic or Bohairic Version, current in Lower or Northern Egypt, and

  2. the Thebaic or Sahidic Version, current in Upper or Southern Egypt.

Neither version is complete.
Of the Bohairic, the Pentateuch, Psalms and Prophets have been published, and other fragments are known. The Sahidic exists in very considerable fragments, which have been much increased by recent discoveries. The British Museum alone has acquired a complete MS. of Deuteronomy and Jonah (with Acts), of the fourth century, a seventh-century palimpsest of Joshua, Judges, Ruth, Judith and Esther, sixty-two leaves of Proverbs, Ecclesiastes, Song of Solomon, Wisdom and Ecclesiasticus, of the same date, and a complete Psalter, also of the seventh century. Mr. Pierpont Morgan has MSS. of I and 2 Kingdoms, Leviticus-Deuteronomy, and Isaiah; and there are other valuable fragments elsewhere. One portion of the Sahidic version is of especial interest; for copies of the book of Job in this version have been discovered containing a text which bears every mark of being its original form. It is shorter than the received text by about one-sixth, omitting in all about 376 verses; but the passages which disappear are in many cases inconsistent with the general argument of the book, and appear to have been inserted by Jewish scholars who did not understand, or did not approve of, the plan of the poem as it was originally written. Indeed the whole Sahidic Old Testament seems to have been at first free from Hexaplar additions, but to have been subsequently revised from MSS. containing these additions, presumably copies of the Hesychian text which was current in Egypt. The Sahidic version was probably made before the end of the second century, the Bohairic somewhat later. Of the third version,

  1. the Middle Egyptian, only a few fragments have as yet been discovered.


 The Ethiopic Version.

Ethiopic Octateuch

With the versions of Egypt may naturally go the version of Ethiopia; but it will require only a brief notice. The Ethiopian manuscripts (most of which were acquired by the British Museum at the time of the Abyssinian war in 1867) are of very late date, but the original translation was probably made in the fourth century after Christ. This version was, no doubt, made from the Septuagint; but it has been questioned whether the extant MSS. really represent this translation, or a much later one, made in the fourteenth century from the Arabic or Coptic. The fact is that at present little can be said to be known about the version at all. Both Old and New Testament are preserved to us entire, though in very late manuscripts, but they have never been properly edited. One special feature, however, of the Ethiopic Old Testament deserves to be noticed. Besides the ordinary books contained in the Septuagint, it includes also two apocryphal books, which have no place in either our Old Testament or our Apocrypha - namely, the book of Jubilees and the book of Enoch. The latter book is of special interest, from its having been quoted in the Epistle of Jude; but it was wholly lost, except for some extracts in Syncellus, until James Bruce brought back some manuscripts of it from Abyssinia in 1773, from one of which it was edited by Archbishop Laurence in 1821. The original Greek remained unknown until 1886, when a little vellum volume was discovered at Akhmim in Egypt, containing the first thirty-six chapters, along with portions of the Gospel and Apocalypse of Peter. Still more recently, the last eleven chapters have been recovered from one of the papyri in Mr. Chester Beatty's collection. The native Abyssinian Church has just produced a new edition of the Ethiopic Bible, with the modern Amharic text in parallel columns, but this is not a critical edition.

The remaining Oriental versions may be dismissed in a few words. A few fragments remain of the Gothic version, made for the Goths in the fourth century by their bishop, Ulfilas, while they were still settled in Moesia, the modern Serbia and Bulgaria. Its chief interest lies in the fact that it was taken from a copy of the Lucianic edition of the Septuagint. 

The Armenian, Arabic, Georgian, and Slavonic versions were all made from the Septuagint, but they have been little studied.

4. The Latin Versions.

Old Latin | The Vulgate | Jerome | - his Psalters | - his OT | Reception of his Version | Its Character

(a) The Old Latin Version.

When Christianity reached Rome, the Church which was founded there was at first more Greek than Latin. St. Paul wrote to it in Greek, the names of most of its members, so far as we know them, are Greek, and its earliest bishops were Greek: one of them, Clement, wrote an epistle to the Corinthians in Greek which is found along with the books of the New Testament in one of the earliest Greek Bibles, the Codex Alexandrinus. There was therefore at first no necessity for a Latin version of the Scriptures; and the necessity, when it arose, was felt less in Rome itself than in the Roman province of Africa. It is in this province, consisting of the habitable part of northern Africa, lying along the southern coast of the Mediterranean, that a Latin Bible first makes its appearance. 

The importance of the Old Latin version, as it is called, to distinguish it from the later version of St. Jerome, is much greater in the New Testament than in the Old. In the former, it is one of the earliest translations of the original Greek which we possess, and is an important evidence for the state of the text in the second century. In the latter it is only a version of a version, being made from the Septuagint, not from the original Hebrew. Historically, moreover, it is of less importance; for it was almost entirely superseded by the version of Jerome, and it exists to-day only in fragments. No entire manuscript survives of the Old Testament in this version; a few books only, and those chiefly of the Apocrypha, exist complete; for the rest we are indebted for most of our knowledge of this version to the quotations in the early Latin Fathers. 

The Old Latin version of the New Testament was extant in Africa in the second century after Christ, and it is probable that the translation of the Old Testament was made at the same time, since it is almost certain that a complete Latin Bible was known to Tertullian (about AD200). Whether the first translation was actually made in Africa it is impossible to say, for want of positive evidence; but this view is commonly held and is at least probable. What is certain is that the version exists in two different forms, known, from the regions in which they circulated, as the African and the European. How far they are independent is uncertain. The original translation was rough and somewhat free; in the European edition the roughnesses are toned down and the translation revised with reference to the Greek. As the translation was originally made before the time of the various editions of Origen, Lucian, and Hesychius, its evidence, wherever we possess it, is useful as a means to the recovery of the earlier form of the Septuagint; and it is observable that its text is akin to that which appears in the Codex Alexandrinus, which seems to indicate an Egyptian origin. Unfortunately it is available only to a limited extent. The apocryphal books of Esdras, Wisdom, Ecclesiasticus, Baruch, and 1 and 2 Maccabees, together with the additions to Daniel and Esther, were not translated or revised by Jerome, and consequently the Old Latin versions of these books were incorporated in the later Latin Bible and remain there to this day. [The Old Latin version of Ecclesiasticus enables us to correct a disarrangement which has taken place in the text of the Septuagint.  In the Greek version, chap.xxx.25-xxi.130 is placed after chap.xxxvi.16a, which is plainly wrong. The Latin version has preserved the true order, which has been followed in our Authorised Version.] The Psalter survives in a very slightly altered form, as explained below; but the historical and prophetical books have disappeared almost completely. The Octateuch is in better case. There has long been a fine manuscript of the fifth century at Lyons, containing portions of Genesis, Exodus and Leviticus, the whole of Numbers, and the first ten chapters of Deuteronomy. To this M. Delisle, Director of the Bibliotheque Nationale in Paris, was able to add in 1895 the rest of Deuteronomy, the whole of Joshua, and Judges as far as xi.21. Probably Ruth was originally included, so that the whole MS. would have been an Octateuch. Ruth has come down in another MS. at Madrid; and Esther, Judith and Tobit are also preserved in that MS. and in others. For the rest we are dependent on a few fragments and quotations in the Fathers.

(b) The Vulgate.

It is very different when we come to the great work of St. Jerome, which, in the main, continues to be the Bible of the Roman Church to this day. Its origin is known to us from the letters and prefaces of its author; its evidence is preserved to us in hundreds and even thousands of manuscripts of all ages from the fourth century to the fifteenth. Its historical importance is enormous, especially for the Churches of Western Europe; for, as we shall see in the progress of our story, it was the Bible of these Churches, including our own Church of England, until the time of the Reformation. We shall have to trace its history in the later chapters of this book; for the present we are concerned with the story of its birth. 

By the end of the fourth century the imperfections of the Old Latin version had become evident to the leaders of the Roman Church. Not only was the translation taken from the Greek of the Septuagint, instead of the original Hebrew, but the current copies of it were grossly disfigured by corruptions. The inevitable mistakes of copyists, the omissions and interpolations of accident or design, the freedom with which early translators handled the text of their original, the alterations of revisers, and the different origin of the African and European forms of the version, all contributed to produce a state of confusion and distortion intolerable to an educated Churchman.


Hence about the year 382 Pope Damasus appealed to the most capable Biblical scholar then living, Eusebius Hieronymus, whom we know better under the abbreviated form of his name, Jerome, to undertake a revision of the Latin Bible. Jerome was born in 346, a native of Stridon in Pannonia, not far from the modern Trieste. Throughout his life he was devoted to Biblical studies. In 374 he set himself to learn Hebrew, then a very rare accomplishment in the West, taking as his teacher a converted Jew. His first Biblical undertaking, however, was not connected with his Hebrew studies. The existing Latin Bible was a translation from the Greek throughout, in the Old Testament as well as in the New, and all that Pope Damasus now invited Jerome to do was to revise this translation with reference to the Greek. He began with the Gospels, of which we shall have to speak later; but about the same time he also made his first revision of the Psalter.

...his Three Psalters.

He produced eventually no less than three versions of the Psalms, all of which are still extant. The first was this very slight revision of the Old Latin version, with reference to the Septuagint, and is known as the Roman Psalter; it was officially adopted by Pope Damasus, and still remains in use in the cathedral of St. Peter at Rome. The second, made between 387 and 390, was a more thorough revision, still with reference to the Septuagint; but Jerome attempted to bring it into closer conformity with the Hebrew by using Origen's Hexaplar text and reproducing his asterisks and obeli; this version was first adopted in Gaul, whence it is known as the Gallican Psalter, and it has held its place as the Psalter in general use in the Roman Church and in the Roman Bible from that day to this, in spite of the superior accuracy of the third version which Jerome subsequently published. This is known as the Hebrew Psalter, being an entirely fresh translation from the original Hebrew. It is found in a fair number of manuscripts of the Vulgate, often in parallel columns with the Gallican version, but it never attained to general usage or popularity.

...his Old Testament.

About the time when Jerome produced his Gallican Psalter, he also revised some of the other books of the Old Testament, such as Job (which alone now survives in this form), with reference to the Hexaplar text; but it would appear that this undertaking was not carried to completion. It is probable that Jerome, as his knowledge of Hebrew increased, grew dissatisfied with the task of merely revising the Old Latin translation with reference to a text which itself was only a translation. He had completed the revision of the New Testament on these lines; but with the Old Testament he resolved to take in hand an altogether new translation from the Hebrew. He appears to have felt no doubt as to the superiority of the Hebrew text over the Greek, and in all cases of divergence regarded the Hebrew as alone correct. This great work occupied him from about the year 390 to 404; and separate books or groups of books were published as they were completed. The first to appear were the books of Samuel and Kings, next the Prophets, then Ezra, Nehemiah and Genesis, then (after an interval) the books of Solomon, and finally the rest of the Octateuch and Esther.

Reception of His Version.

In the prefatory letters prefixed to these books, Jerome tells us much of his work and its reception. In spite of much individual support which he received, the general attitude towards it was one of great hostility. The sweeping nature of the changes introduced, the marked difference in the text translated, alienated those who had been brought up to know and to love the old version, and who could not understand the critical reasons for the alteration. Jerome felt this opposition keenly, and raged against what he regarded as its unreasonableness; and his sensitiveness, not to say irritability, finds vigorous expression in his prefaces. We who have seen the introduction of a Revised Bible in our own country, intended to supersede the version to which England has been devotedly attached for centuries, can understand the difficulties which surrounded the work of Jerome. Gradually, as we shall see in a later chapter, the superior accuracy and scholarship of his version gave it the victory, though not in a perfect or complete form. The Gallican Psalter continued to hold its own, and was never replaced by the version from the Hebrew. The apocryphal books he wished to reject entirely, because they found no place in the current Hebrew Bible. He did indeed consent reluctantly to make a very hurried translation of the books of Judith and Tobit; but the remaining books he left untouched. In spite of this, they continued to find a place in the Latin Bible; and the Vulgate, as finally adopted by the Roman Church, contains these books in the form in which they had stood, before the days of Jerome, in the Old Latin version. In the rest of the Old Testament, Jerome's version ultimately superseded the Old Latin, and in the New Testament his revision of the Old Latin held its ground. To this composite Bible, consisting partly of unrevised translations from the Greek, partly of revised translations from the same, and partly of translations from the Hebrew, was given in later days, when it had been generally accepted in Western Europe, the name of the "Vulgate," or commonly received translation; and of this, the Bible of our own country until the Reformation, and of the Roman Church until to-day, we shall have much to say hereafter as we trace its history through the centuries. We shall also reserve for later chapters an account of the chief manuscripts in which it is now preserved. In the present chapter we have to do with it only as it affords evidence which may help us to recover the original Hebrew text of the Old Testament.

Its Character.

In this respect its importance is not to be compared with that of the Septuagint. The Hebrew text accessible to Jerome was practically identical with that which is accessible to ourselves; for although the Massoretes themselves are later in date than Jerome by several centuries, yet, as we have seen, the text which they stereotyped had come down practically unchanged since the beginning of the second century after Christ. Hence the version of Jerome is of little help to us in our attempt to recover the Hebrew text as it existed in the centuries before the Christian era; on the other hand, if the Massoretic text is in itself superior to the Greek version as a whole, then the Vulgate is a more satisfactory national Bible than the Septuagint. The translation itself is of unequal merit; some parts are free to the verge of paraphrase, others are so literal as to be nearly unintelligible; but on the whole the work is one of very great merit, and justifies the commanding position which Jerome holds among the Fathers of the Roman Church. Jerome was, indeed, for the West what Origen was for the East - the greatest Biblical scholar which the Church produced before the revival of learning at the end of the Middle Ages.

5. Condition of the Old Testament Text.

Most Versions too late to Help | Evidence of the Samaritan Pentateuch | LXX v Massoretic | Hebrew Text sure to be corrupt | - certainly in some places | LXX not always Trustworthy | LXX Additions | Ecclesiasticus- Hebrew Text | Minor Corruptions | Deliberate Falsification of Hebrew not Proven | Summing-up. 

The Vulgate is the last of the versions of the Old Testament which need be mentioned here; and now we come back to the question with which we ended the preceding chapter. What light, after all, do these versions throw on the text of the Old Testament? Do they help us to get behind the Massoretic text, and see what the words of the Scriptures were when they were first written down? And, if so, does this earlier evidence confirm the accuracy of the Massoretic text, or does it throw doubt upon it? With the answer to this question we can close our examination of the Old Testament text. 

A diagram may serve to summarise, in broad outline, the information which has been given above.

Original Text    
400 Sam.  
200   Sept.  
100       Massoretic  
200     Old Latin Syriac      
300     Coptic    
400     Ethiopic Vulgate    
Samaritan Septuagint Hebrew   Targums


Most of the Versions too Late to Help Us.

In the first place it will be clear that some of the versions we have described must be excluded on the ground that they are not translations of the Hebrew at all. Thus the Coptic, Ethiopic, Gothic, Armenian, Arabic, Georgian, Slavonic, and Old Latin versions were made from the Greek of the Septuagint; and they can only indirectly help us to recover the original Hebrew. Their value is that they help us to restore the original text of the Septuagint; and from the Septuagint we may get on to the Hebrew. In the next place, the Peshitta Syriac and the Latin Vulgate, though translated from the Hebrew, were translated at a time when the Hebrew text was practically fixed in the form in which we now have it. The Peshitta was made in the second or third century, the Vulgate at the end of the fourth; but we have already seen that we can trace back the Massoretic text to about the beginning of the second century. In some cases, when the Hebrew has been corrupted at a comparatively late date, these versions may show us the mistake; but their main value arises from the fact that, at the time when they were made, the Hebrew vowel-points were not yet written down, but were supplied in reading the Scriptures according to the tradition current among the Jews. Hence the Peshitta and the Vulgate show us in what way the absent vowels were supplied at a date very much earlier than any of our existing manuscripts. The same is the case with the Greek versions of Aquila, Theodotion, and Symmachus. They were made from the Hebrew, but from a Hebrew text too late to be of much service to us in our present inquiry.

Evidence of the Samaritan Pentateuch.

There remain the Samaritan and the Septuagint versions. Of these the Samaritan is the oldest; and as it is not really a translation into a different language, but a direct descendant of the original Scriptures in the same language and written in the same characters, its evidence might be expected to be of exceptional value. Unfortunately, however, it relates only to the Pentateuch; and we have seen (p.51) that it is exactly here that help is least required, and that the variations of the Samaritan text, even where they appear to be right, are not of very great or striking importance. With the Septuagint it is quite otherwise. It contains all the books of the Old Testament, including those which the Jews finally refused to accept as inspired; and its variations are, in many of the books, both numerous and important. The real question to be debated, then, is this: Does the Septuagint or the Massoretic text represent most accurately the words and form of the Old Testament Scriptures as they were originally written?

Septuagint v. Massoretic.

So far as the weight of authority goes, the preponderance is decidedly in favour of the Hebrew. Origen and Jerome, the two greatest Biblical scholars of antiquity, deliberately abandoned the original Septuagint and its descendants, the translations made from it, in order to produce versions which should correspond as nearly as possible with the Hebrew. So, too, in the modern world, all the translators of the Bible whose scholarship was equal to it went to the Hebrew for their text of the Old Testament, while those who could not read Hebrew fell back upon the Vulgate, which was itself translated from the Hebrew. Our own Authorised and Revised Bibles, as well as nearly all the translations which preceded them, rest almost entirely upon the Massoretic text, and only very rarely follow the versions in preference to it. And this is very natural; for the Old Testament books were written in Hebrew, and it seems reasonable to suppose that they would be best represented in the Hebrew manuscripts. In the case of no other book in the world should we look to a translation rather than to copies in the original language for the best representation of the contents of the work. Since the last century, however, there have been scholars who have maintained that the Septuagint, the origin of which goes back to a date far earlier than that to which the Massoretic text can be traced, comes nearer to the original Hebrew than do the Hebrew manuscripts of the Massoretic family. It would be absurd to attempt to decide the point authoritatively in such a work as this; but the conditions of the problem can be stated, and the apparent course of the controversy indicated in brief.

The Hebrew Text sure to be Corrupted.

In the first place it is only natural that the Hebrew text should have suffered considerable corruption. If we take the year 100 after Christ as representing the date to which we can trace back the existence of the Massoretic text, there is still a gap of many centuries before we reach the dates at which most of the books were composed. Nearly a thousand years separate us from the earliest of the Prophets, and even if we accept the latest date which modern criticism assigns to the composition of the Pentateuch in its present form, there are still more than five hundred years to be accounted for. It would be contrary to reason to suppose that the text had been handed down through all these centuries without suffering damage from the errors of scribes, the alterations of correctors, or the revision of editors, especially when we remember that in the course of that period the whole style of writing had been changed by the introduction of the square Hebrew characters, that the words were not divided from one another, and that the vowels were not yet indicated by any marks. It is thus natural in itself that the Hebrew text as we have it now should need some correction. It is also natural that the Septuagint version, which we can trace back to an origin more than 350 years earlier than the Massoretic text, should in some cases enable us to supply the needed correction. The text of the Septuagint may itself have suffered much corruption between the time of its composition and the time to which our direct knowledge of it goes back; but it is contrary to reason to suppose that it has always been corrupted in those places where the Hebrew has been corrupted, and that it does not sometimes preserve the right reading where the Hebrew is wrong.

And certainly Corrupt in Some Places.

A partial confirmation of this conclusion is provided by the Targums, the earliest portions of which go back a century or more before the formation of the Massoretic text. In these there are indications that the text on which they are based, though very like the Massoretic text, was not identical with it. We can, however, go further, and show that there is a much larger number of passages in which corruption has almost certainly taken place between the date at which the Septuagint was written and that at which the Massoretic text was formed. It would need an entire treatise to do this thoroughly, but the reader of the Variorum Bible will find a considerable number of places noted in which the reading of the Septuagint makes better sense than that of the Hebrew. In not a few passages the Hebrew gives no natural meaning at all; for instance, Ex.xiv.20; 1 Sam.i.21; xxvii.10 (where even the Authorised Version departs from the Massoretic text); much of 1 Kings vi. and vii.; Job iii.14; xxxv.15, and many other passages indicated in the Variorum Bible. In other places verses are supplied by the Septuagint which are not in the Hebrew; in these it will be a matter for critics to decide in each case whether the Hebrew has wrongly omitted words, or the Septuagint wrongly inserted them, but it is not likely that the answer will always be the same. A list of some such passages has already been given on p.79. Again, take the larger variations there mentioned in the books of Jeremiah and Job. In the former the arrangement found in the Septuagint is by many scholars considered preferable to that of the Hebrew, and its text in many doubtful passages appears to be superior. In Job the proof is even more complete; for a large number of passages in it, which had already been believed, on the ground of their style, to be later additions to the Hebrew, have recently been shown to have been absent from the original text of the Septuagint, and to have been added by Origen in his Hexapla, with the usual marks indicating that they had been introduced by him from the Hebrew. Once more, in the Pentateuch we find the Septuagint and the Samaritan version often agreeing in opposition to the Hebrew; and since there is no reasonable ground for asserting that either of these translations was influenced by the other, we can only suppose that in such passages they represent the original reading of the Hebrew, and that the Massoretic text is corrupt.
To this it may be added that the "Book of Jubilees," a Jewish work written not long before the fall of Jerusalem (ad 70) and containing a modified version of the story of Genesis, frequently supports the Septuagint and Samaritan readings in preference to those of the Hebrew.

But the Septuagint not always Trustworthy.

It seems, then, reasonable to conclude that in many cases the Septuagint contains a better text than the Hebrew; and if this is so, it is likely that it is often right in passages where we are not able to decide with certainty between alternative readings. Can we go further and say that it is generally so, and that wherever the two differ, the presumption is in favour of the Septuagint? Certainly not, without considerable qualifications. There can be no doubt, first, that the Septuagint as originally written contained many mistakes; and, secondly, that the text of it has been much corrupted in the earlier course of its history. It must be remembered that the Septuagint was translated from a Hebrew text in which the words were not separated from one another and were unprovided with vowel-points. Hence some of the differences between the Septuagint and the Hebrew do not imply a difference of reading at all, but simply a difference in the division of the letters into words or in the vowel points supplied. Sometimes the one may be right and sometimes the other; but in any case the difference is one of interpretation, not of text. Then, again, there can be no doubt that the authors of the Septuagint made many actual mistakes of translation. Hebrew, it must be remembered, was not their habitual language of conversation; it was a matter of study, as old English is to scholars to-day, and it was quite possible for them to mistake the meaning of a word, or to confuse words which were written or spoken nearly alike. The possibility of such mistakes must be borne in mind, and only a good Hebrew scholar can warn us of them.
[Some interesting examples of errors caused by the Greek translator having misunderstood the Hebrew, or having supplied the wrong vowel-points, are given in the preface to the Variorum Apocrypha.]

Additions in Septuagint.

It is a more difficult point to decide whether the authors of the Septuagint made deliberate additions to the text. Translators held a different view of their rights and duties from that which would be accepted to-day. They thought themselves at liberty to add explanatory words and phrases, to paraphrase instead of adhering closely to their original, to supplement what they believed to be omissions (often by incorporating words from other passages where the same or similar events were recorded, as from Kings into Chronicles, and vice versa), even to omit passages which they regarded as unnecessary or unedifying, or insert incidents which they believed to be true and edifying. This would seem to be the case with the additions to the books of Daniel and Esther, which the Jews refused to accept as part of the inspired Scriptures, and which have been banished to the Apocrypha in the English Bible. In smaller details, the authors of the Septuagint seem at times to have softened down strong expressions of the Hebrew, no doubt from a feeling that the more refined literary taste of Alexandria would be offended by them.

The Hebrew Text of Ecclesiasticus.

A welcome and valuable contribution to our comprehension of the relation between the Septuagint and the Massoretic Hebrew was made in 1897 by the publication of a portion of the Hebrew original of the book of Ecclesiasticus, previously believed to be wholly lost. The Hebrew text was known to Jerome, and there is evidence that it was still in existence early in the tenth century; but thenceforward, for a space of more than 950 years, no traces of it could be met with. In 1896, however, Mrs. Lewis, the fortunate discoverer of the Sinaitic Syriac manuscript of the Gospels, brought back from the East a single leaf, which, on being examined at Cambridge, was found to contain part of the original Hebrew text of Ecclesiasticus; and almost simultaneously Dr. Ad. Neubauer at Oxford, in examining a mass of fragments sent to England by Professor Sayce, discovered nine more leaves of the same MS., following immediately after the Cambridge leaf. The total amount of text thus recovered includes chapters xxxix.15-xlix.11; and the whole was edited by Mr. Cowley and Dr. Neubauer, of the Bodleian Library, Oxford. [A very convenient small edition was issued in 1898 for those who are not Hebrew scholars, giving a translation of the Hebrew side by side with the Revised Version of the same portion of the book. A short introduction supplies all the necessary information.] The manuscript is on paper, and was written about the end of the eleventh or beginning of the twelfth century. 

The most striking feature about the discovery is the extent of the divergence between the Hebrew and the Greek versions; and the character of the divergence shows that it is generally due to the mistakes or omissions of the Greek translator. It is a most instructive exercise to read the newly recovered original side by side with the notes in the Variorum Apocrypha, which indicate the passages previously suspected of error in the Greek, the variations found in the other versions, and the conjectures of editors. Sometimes the suspicions of scholars are confirmed; often it is seen that they could not go far enough, nor divine the extent to which the Greek departed from the original. A small instance may be given here, from Ecclus.xl.18-20:





(From the Revised Version of 1895)



The life of one that laboureth,
and is contented, shall be made sweet;

 A life of wine and strong drink is sweet,


And he that findeth a treasure is above both.

But he that findeth a treasure is above them both.


Children and the building of a city establish a man's name;

A child and a city establish a name,


But he that findeth wisdom is above them both.


Offspring (of cattle) and planting make a name to flourish,


And a blameless wife is counted above both.

But a woman beloved is above them both. 


Wine and music rejoice the heart;

 Wine and strong drink cause the heart to exult,


And the love of wisdom is above both.

But the love of lovers is above them both.

The divergences in verses 18 and 20 are evidently due to a desire to improve the sentiments of the original by removing the laudatory mention of "strong drink," and the substitution of "the love of wisdom" for "the love of lovers"; while the omission in verse 19, whether it be accidental or intentional, distorts the sense of the passage. That the Hebrew text is the more authentic cannot be questioned; and this is but a sample of what is found throughout the book. It is clear, both that the translator took considerable liberty of paraphrase, and that he sometimes did not understand the Hebrew before him. This latter fact might seem strange, since we know (from the translator's preface) that the original was probably written about 200-170 BC, and the translation (by the author's grandson) in 132 BC, so that the interval of time between them was short; but it is accounted for both by the fact that the translator was no scholar, and by the transition through which the Hebrew language passed during this period. Classical Hebrew, the language of nearly all the canonical books of the Old Testament, was passing into modern or Rabbinical Hebrew, a change quite sufficient to disconcert a moderate scholar. The Rabbinical element appears already in the book of Ecclesiastes; and hitherto it has been supposed that in Ecclesiasticus, which is probably of somewhat later date, it would be more strongly developed. The newly discovered manuscript, however, shows that Jesus Ben-Sira wrote in pure classical Hebrew, equal to that of the Psalms; and no doubt it is partly to this cause that the errors of the translator are due. The moral to be drawn from this discovery is consequently one of caution in assuming that variations (even considerable ones) in the Septuagint from the Massoretic Hebrew necessarily imply a different original text. They may do so, no doubt; but we must be prepared to make considerable allowances for liberty of paraphrase and for actual mistakes, especially in the case of the books which are likely to have been the latest to be translated. When the earliest parts of the Septuagint were translated, a competent knowledge of classical Hebrew must have been much commoner, and a higher standard of accuracy, though not necessarily of literalness, may be expected.

Minor Corruptions.

As to the minor corruptions of the Septuagint text, the history of it in the preceding pages explains these sufficiently. It is no easy task, in many places, to be sure what the true reading of the Septuagint is. Some manuscripts represent the text of Origen, in which everything has been brought into conformity with the Hebrew as it was in his day; many are more or less influenced by his text, or by the versions of Aquila and Theodotion. Some represent the edition of Lucian; others that of Hesychius. Even those which belong to none of these classes do not agree among themselves. The great manuscripts known as A and B frequently differ very markedly from one another, and X sometimes stands quite apart from both. It is clear that in many cases it is impossible to correct the Hebrew from the Greek until we have first made sure what the Greek reading really is.

Deliberate Falsification of Hebrew not Proven.

One further possibility remains to be considered, that of deliberate falsification of either Greek or Hebrew for party purposes. Such accusations were made, both by Christians and by Jews, in the early centuries of the Church's history, when the Jews held to the Hebrew text as it was fixed about AD100, and the Christians to the Septuagint. They have been renewed from time to time; and a modern controversialist, Sir H. Howorth, in his contention for the superiority of the Septuagint, has declared the Massoretic text to have been deliberately altered by the Jews with an anti-Christian purpose. But the proof for so serious a charge is wholly lacking. It is true that the Hebrew Bible as we know it assumed its present form at a time when the antagonism between Jew and Christian was strongly marked, and probably under the direction of the Rabbi Akiba, the great leader of the extreme party of the Jews at the end of the first century. At such a time and under such a leader it might seem not impossible that an attempt would be made to remove from the Old Testament those passages and expressions to which the Christians referred most triumphantly as prophecies of Christ. The best answer to such a charge is that these passages have not been removed, and that the differences between the Massoretic text and the Septuagint are by no means of this character. Nothing can have been gained, from the party point of view, by altering the order of the prophecies of Jeremiah, or by expanding the book of Job. The books of Wisdom and Ecclesiasticus, which were ejected from the Hebrew text and retained in the Greek, do not testify of Christ more than the undisputed books which remain in both. The Christians had less reason to feel special interest in the books of the Maccabees than the patriotic Jews. Indeed, it is untrue to say that the books of the Apocrypha were at this time ejected from the Hebrew Bible; the fact being that they had never formed part of it, and were never quoted or used on the same level as the books recognised as inspired. It is true that one verse has dropped out of a long list of towns (after Josh.xv.59), in which was contained (as the Septuagint shows; see Variorum footnote) the name of "Ephratah, which is Bethlehem," by the help of which the reference to Ephratah in Psalm cxx.6 might be interpreted as a prophecy of our Lord's birth at Bethlehem; but seeing that the same identification is repeated in four other places, including the much more strongly Messianic passage in Micah v.2, the omission in Joshua alone would be perfectly useless for party purposes, and may much more fairly be explained as an accident. It is needless to add that the greater prophecies of the Messiah, such as the fifty-third chapter of Isaiah, stand quite untouched in the Hebrew, and that the vast majority of the differences between the Hebrew and the Greek throughout the Old Testament could have no possible partisan motive whatever.


The authors of our Revised Version of the Old Testament, while recognising the probable existence of earlier editions of the Hebrew differing from the Massoretic text, yet declare that "the state of knowledge on the subject is not at present such as to justify any attempt at an entire reconstruction of the text on the authority of the versions," and have consequently" thought it most prudent to adopt the Massoretic Text as the basis of their work, and to depart from it, as the Authorised Translators had done, only in exceptional cases." There can be no doubt that they did rightly. The versions have as yet been too insufficiently studied to justify a general use or a rash reliance upon them. When the text of the Septuagint, in particular, has been placed on a satisfactory footing (to which it is to be hoped the large Cambridge edition will greatly contribute) it will be time enough to consider how far its readings may be taken in preference to those of the Hebrew. It is probable that eventually a much fuller use will be made of the Septuagint than has hitherto been the case, and those have done good work who have called attention, even in exaggerated tones, to the claims of the ancient Greek version; but no general substitution of the Greek for the Hebrew as the prime authority for the text of the Old Testament will be possible unless the universal assent of students be won to the change. It will not be enough for one section of specialists to take up the cry, and, proclaiming themselves to be the only advanced and unprejudiced school, look down upon all others as unenlightened laggards. Such schools and such cries, stimulative as they are of thought and of work, are for the moment only. If the Massoretic text is ever to be driven from the assured position of supremacy which it has held since the days of Origen and of Jerome, it will only be when the great bulk of sober criticism and the general intelligence of Biblical students have been convinced that the change is necessary. It is very doubtful whether such a conviction will ever be reached; but it is probable that increasing use will be made of the Septuagint evidence, and students will do well to keep an eye on it in their work on the Old Testament.