V. Other Greek versions, p. 75.

VI. The Venetian Greek, p. 76.

VII. The Peshito, or old Syriac, p. 77., from which are derived1. The Karkaphensian, p. 81.

2. Arabic versions, p. 85.

VIII. Some Arabic translations, as that of Saadias Gaon, p. 84. ; the Pentateuch, published by Erpenius, p. 84. ; the version of Joshua, &c., in the Polyglotts, p. 84. ; and the Genesis, &c., by Saadias Ben Levi Asnekot, p. 84.

IX. The Vulgate, p. 89., from which there are —

1. Some Arabic versions, p. 85.

2. Persian of the Psalms, p. 86.

3. Anglo-Saxon, p. 94.

X. Persian translations of the Pentateuch, and that of Proverbs, p. 86.

From the Samaritan Pentateuch there are

1. The Samaritan version, p. 33.

2. Arabic, p. 33.

3. Greek, p. 76.]

On the application of ancient versions to the ascertaining of various readings, and on the benefit which may be derived from them in the interpretation of the Scriptures, remarks will be hereafter made.

SECTION III

ON THE VARIOUS READINGS OCCURRING IN THE SCRIPTURES.

§ 1. On the Causes of Various Readings.

I. The Christian faith not affected by what are called various readings.— II. Nature of various readings - Difference between them and mere errata.-III. Notice of the principal collations and collections of various readings. IV. Causes of various readings:-1. The negligence or mistakes of transcribers ;-2. Errors or imperfections in the manuscript copied; 3. Critical conjecture;-4. Wilful corruptions of a manuscript from party-motives.

I. THE Old and New Testaments, in common with all other ancient writings, being preserved and diffused by transcription, the admission of mistakes was unavoidable; which, increasing with the multitude of copies, necessarily produced a great variety of different readings. Hence the labours of learned men have been directed to the collation of manuscripts, with a view to ascertain the genuine reading; and the result of their researches has shown that these variations are not such as to affect our faith or practice in any thing material: they are mostly of a minute, and sometimes of a trifling nature. "The real text of the sacred writers does not now (since the originals have been so long lost) lie in any single manuscript or edition, but is dispersed in them all. It is competently exact indeed, even in the worst

1

manuscript now extant; nor is one article of faith or moral precept either perverted or lost in them." It is therefore a very ungrounded fear that the number of various readings, particularly in the New Testament, may diminish the certainty of the Christian religion. The probability, Michaelis remarks, of restoring the genuine text of any author, increases with the increase of the copies; and the most inaccurate and mutilated editions of ancient writers are precisely those of whose works the fewest manuscripts remain. Above all, in the New Testament, the various readings show that there could have been no collusion; but that the manuscripts were written independently of each other, by persons separated by distance of time, remoteness of place, and diversity of opinions. This extensive independency of manuscripts on each other is the effectual check of wilful alteration; which must have ever been immediately corrected by the agreement of copies from various and distant regions out of the reach of the interpolator. By far the greatest number of various readings relate to trifles, many of which cannot be made apparent in a translation; and, of the rest, very few produce any alteration in the meaning of a sentence, still less in the purport of a whole paragraph.

II. However plain the meaning of the term "Various Reading may be, considerable difference has existed among learned men concerning its nature. Some have allowed the name only to such readings as may possibly have proceeded from the author; but this restriction is improper. Michaelis's distinction between mere errata and various readings appears to be the true one. "Among two or more different readings, one only can be the true reading; and the rest must be either wilful corruptions or mistakes of the copyist." It is often difficult to distinguish the genuine from the spurious; and, whenever the smallest doubt can be entertained, they all receive the name of various readings; but, in cases where the transcriber has evidently written falsely, they receive the name of errata.

III. Human life is too short to allow of a thorough examination of all those monuments which are indispensably necessary to sacred criticism, in addition to the many other subjects which are equally worthy of attention. But, as many learned men have from time to time investigated different documents, extensive collections of various

Dr. Bentley's Remarks on Free-thinking, Rem. xxxii. (Bp. Randolph's Enchiridion Theologicum, edit. 1812, vol. ii. p. 408.) The various readings that affect doctrines, and require caution, are extremely few, and easily distinguished by critical rules; and, where they do affect a doctrine, other passages confirm and establish it. See examples of this observation in Michaelis, vol. i. p. 266., and Dr. Nares's Strictures on the Unitarian Version of the New Testament, pp. 219-221.

2

* Michaelis, Introduction to the New Testament, vol. i. chap. vi. pp. 263-268. "In profane authors," says Dr. Bentley, "(as they are called) whereof one manuscript only had the luck to be preserved - as Velleius Paterculus among the Latins, and Hesychius among the Greeks the faults of the scribes are found so numerous, and the defects so beyond all redress, that, notwithstanding the pains of the learnedest and acutest critics for two whole centuries, those books still are, and are likely to continue, a mere heap of errors. On the contrary, where the copies of any author are numerous, though the various readings always increase in proportion, there the text, by an accurate collation of them made by skilful and judicious hands, is ever the more correct, and comes nearer to the trne words of the author." Remarks on Free-thinking, in Enchirid. Theol. vol. ii. p. 406.

readings have gradually been formed, of which the critic should avail himself.

With regard to the Old Testament, some beginnings were made by those ancient Jews to whom we owe the rejections and corrections of the scribes, and other observations, already noticed in pp. 22, 23, 26, and 27. of this volume. More recently the rabbis Todrosi, Menahem, and Norzi, collected a larger apparatus.' Sebastian Munster was the first Christian editor, who in 1536 added some various readings. Not many more are found in Vander Hooght's edition, printed 1705; but in the subsequent editions of John Henry Michaelis, in 1720, and of Houbigant in 1753, the critical collation of various readings was very considerably enlarged. At length, after many years of unremitting toil, Dr. Kennicott produced his edition of the Hebrew Bible, printed at Oxford in 1776-80, which contained various readings collected throughout Europe, from six hundred and thirty-four manuscripts, from forty-three editions, and from both the Talmuds. From this apparatus De Rossi selected the more important readings; and, after collating eight hundred and twentyfive manuscripts and three hundred and seventy-five editions, and examining fully the ancient versions and book of the rabbins, even in manuscript, he published all the various readings he had observed, in four volumes, quarto, in 1784-88, at Parma, to which he added a supplement or scholia, in 1798. As the price of their publications necessarily places them out of the reach of very many biblical students, the reader, who is desirous of availing himself of the results of their laborious and learned researches, will find a compendious abstract of them in Mr. Hamilton's Codex Criticus. (London, 1821,8vo.) For the Septuagint Version, the principal collation of various readings will be found in the edition commenced by Dr. Holmes, and completed by the Rev. Dr. Parsons, at Oxford, in 1798-1827, in five volumes, folio. [Tischendorf, in his edition of 1850, has collected readings from other MSS.]

IV. As all manuscripts were either dictated to copyists or transscribed by them, and as these persons were not supernaturally guarded against the possibility of error, different readings would naturally be produced, 1. By the negligence or mistakes of the transcribers; to which we may add, 2. The existence of errors or imperfections in the manuscripts copied; 3. Critical emendations of the text; and 4. Wilful corruptions made to serve the purposes of a party. Mistakes thus produced in one copy would of course be propagated through all succeeding copies made from it, each of which might likewise have peculiar faults of its own; so that various readings would thus be increased, in proportion to the number of transcripts that were made.

[Various readings have arisen both from negligence and from design-not always in the latter case with an intention of depraving the text, but much more frequently with the well-meaning purpose of improving it.

'An account of their labours is given by Dr. Kennicott in his Dissertatio Generalis, pp. 111-131., and by De Rossi in his Varis Lect. Proleg. pp. XXXIX—XLIII,

Mistakes have sometimes been caused, when the copyist has had the MS. before him, by visual imperfection. Many of the Hebrew letters nearly resemble each other: 2 and 2, and 3, 7 and 7, and П, ↑ and ', ↑ and ↑, and, and ; and hence they have been interchanged; as for example, , Neh. xii. 3.,, xii. 14.; 7, 1 Chron. ix. 15.; 1, Neh. xi. 17. Similar changes are found in Greek MSS. Thus, the Codex Cottonianus of the book of Genesis furnishes, among many others, the following examples: B and M have been interchanged in Gen. xliii. 11., τερεμινθον for τερέβινθον ; I and K, yuvnyos for κuvnyòs, x. 9.: consonants have been added, γυναικαν for γυναῖκα, xi. 31.; or doubled, as Σενααρ for Σεναάρ, x. 10. vowels are interchanged, ηxпν for πx, vi. 16.; or used for diphthongs, as ovdes for ovdeìs, xxxi. 44. Then, since letters were employed as numerals, by the confounding of 2 and we find, 2 Sam. xxiv. 13., 7 years; while 1 Chron. xxi. 12., followed by the Septuagint, has 3 years. From the like cause proceeded a transposition of letters; thus, p, Ezr. ii. 46., 2, Neh. vii. 48.; of words, compare 2 Sam. vi. 2. with 1 Chron. xiii. 6.; of clauses, compare Psal. xcvi. 9-11. with 1 Chron. xiii. 6.; also the omission of letters, words, or clauses, e.g. n, Neh. xi. 5., n, 1 Chron. ix. 15. This was especially likely to happen if two clauses had a similar ending. An example appears in 2 Kings vii. 13.; where the transcriber's eye caught the first, and he in consequence repeated some words which encumber the sense, and do not occur in several MSS., or in the ancient versions, Greek, Latin, and Syriac.

When the MS. was dictated to a copyist, mistakes might arise from imperfect hearing, as 7, 1 Sam. xxii. 18., where Keri,

;

for, 1 Kings xxii. 49. A curious example of this occurs in 1 Sam. xvii. 34. In the Bomberg second edition, was substituted for ; and the error ran through very many editions.

If the copyist did not carefully keep his eye upon his exemplar, mistakes might occur through defect of memory; and thus words or clauses might be transposed, or omitted; or synonymes might be substituted; or the expressions of parallel passages be introduced.

,וַיֹּאמֶר and וַיְדַבֵּר ; .36 .are interchanged, Lev. xxv בַּל and אל ,Thus אֱלֹהִים or אֲדוֹנִי with יְהוָה Kings i. 10. ; and often 2

Then there were errors of judgment as to the proper division of words, e.g. n-hy, Psal. xlviii. 15. for by; the misapprehension of abbreviations, e.g. ' N, Psal. xxv. 37. ; where the Septuagint has Ovμóv pov, equivalent to N. In Psal. xlii. 6, 7., then of v. 7. should be transferred to v. 6., and united with, to be taken from 15. The symmetry of the composition is thus readily restored. Further, the letters termed custodes linearum2 were occasionally taken for a part of a word, as Diy for ww, Isai. xxxv. 1., the of the following word having been written as such a custos. Marginal glosses, also, were sometimes introduced into the text: thus, accord

1 Dr. Holmes's Edition of the Septuagint, vol. i. præf. cap. ii. § 1.

'Words are never divided in Hebrew MSS. When the space at the end of a line is insufficient to contain the next word, one or more letters of it are added, that a blank may not be left; but the whole word is written in the following line. The letters so put in to fill a blank are called custodes linearum.

ing to some,

י.הַלְלוּ־יָה

ns, Isai. vii. 17., and liturgical notes, as

Alterations are said to have been made designedly; and the Jews have been specially accused of wilful corruption. The charge, however, can hardly be substantiated. Had they corrupted the Scriptures before our Lord's time, he would surely have noticed and condemned so great a sin; and after the Christian era corruption was well-nigh impracticable. Accusations of this nature by the early fathers may be accounted for by the fact that they used the Septuagint, and regarded deviation from that a falsifying of the divine word. Jerome, it is true, who was acquainted with the Hebrew original, seems in one place to imagine a Jewish corruption; but his deliberate opinion was utterly opposed to a general charge of the kind. Occasional alterations, no doubt, have been made with the ill-judged design of improving the text; but there is little cause for suspecting any great departure from the original. And, if some critics have gone too far in maintaining the perfect accuracy of the Masoretic text, others have more unwarrantably and more dangerously run to an extreme of opposite error.]

§ 2.

Sources whence the True Readings are to be determined. I. Manuscripts.-II. The most ancient and the best editions.—III. Ancient versions.-IV. The writings of Josephus (for the Old Testament).— V. Parallel passages. —VI. Quotations from the Old and New Testaments in the works of the fathers.-VII. Critical conjecture.

THE causes of various readings being thus ascertained, the next step is to consider the SOURCES WHENCE THE TRUE READING IS TO BE DETERMINED.

The legitimate sources of emendation are, 1. Manuscripts; 2. The most ancient and best editions; 3. Ancient versions; 4. The writings of Josephus, together with the Masorah and the Talmud (for the Old Testament); 5. Parallel passages; 6. Quotations from the Old and New Testaments in the works of the fathers, or in fragments of heretical writings; and, 7. Critical conjecture. But these

See Cappel, Crit. Sacr. (edit. Vogel) tom. i. lib. i. capp. v. et seq.: many of the examples, however, there adduced, will not bear examination; De Wette, Einleitung, §§ 81. &c.; Keil, Einleitung, § 204. For an account of the variations of readings in the New Testament see Dr. Tregelles, vol. iv. chap. vi.

2 ...

incertum habemus, utrum Septuaginta interpretes addiderint, omnis homo, et, in omnibus, an in veteri Hebraico ita fuerit, et postea a Judæis deletum sit. In hanc me autem suspicionem illa res stimulat, quod verbum. apostolus. numquam protulisset nisi in Hebræis voluminibus haberetur." Hieron. Comm. in Epist. ad Galat. lib. ii. cap. iii. v. 10.

....

"Quod si aliquis dixerit, Hebræos libros postea a Judæis esse falsatos, audiat Origenem quid in octavo volumine Explanationum Isaiæ huic respondeat quæstiunculæ, quod num. quam Dominus, et apostoli, qui cætera crimina arguunt in Scribis et Pharisæis, de hoc crimine, quod erat maximum, reticuissent. Sin autem dixerint post adventum Domini Salvatoris, et prædicationem apostolorum libros Hebræos fuisse falsatos, cachinnum tenere non potero, ut Salvator, et evangelistæ, et apostoli ita testimonia protulerint, ut Judæi postea falsaturi erant." Id. Comm. in Isai. Proph. lib. iii. cap. vi. vv. 9. et seqq.