son may be twenty-two years: for D, 42, could easily have arisen from 5, 22, by a mistake of the eye. Comp. also II Sam. xxi. 19., with I Chron. xx. 5., where the former is plainly to be corrected by the latter. It not unfrequently happens, however, that such is the state of parallel places as to make a judgment doubtful.

If parallel places of the same book are at variance, one or other is no doubt erroneous, and the point to be investigated is, which reading could the more easily arise from the other. In determining this, the critic is often at a loss.-But if an author refer to some passage of a book more ancient than himself, in which there is now a various reading, he affords the most ample testimony to the true reading of that passage. Thus Ezekiel, c. xxi. 32, (27.,) is evidence that in his time the reading in Gen. xlix. 10., wash, for he explains it by Dawan is rx, whose right it is.

§ 142. Probability as to what the author has written. When any author has composed a work in a certain country, in a particular place, at a definite time, and with circumstances of a determinate kind, all these characteristics are impressed on his composition, and may be observed by nice attention. From this source arises the rule of criticism, that a reading which is at variance with the country, place, age, and circumstances of an author is spurious. It is evident, therefore, that a knowledge of the history of his author

is necessary for a critic.- -Each author has also his peculiar conformation of language, character of style, mode of thinking, range of figures, and particular modification of sentiment, from all of which also must be deduced critical rules peculiarly adapted to each book. Thus, for example, in Job, Hosea, Micah, Joel, Isaiah, where the style of writing is beautiful and sublime, the elegant reading is the more probable; but in Haggai, Malachi, Ezra, and Nehemiah, where the composition is inferior, the contrary rule holds good. In Isaiah and Micah the reading which contains a paronomasia is to be preferred; in the Psalms of David, the more agreeable; in those of Asaph, and the poems of Moses, the more sublime. But this is a subject which requires great discrimination and attention, and a repeated perusal of the books.

§ 143. General critical rules.

A very extensive and accurate knowledge of very many subjects, some of them very minute, is necessary for a critic. A knowledge of manuscripts, of editions, of versions, of ancient Jewish and Christian writings, of the origin of various readings, of the history of the text and versions, is particularly necessary. The critic moreover ought to be endowed with an acute judgment, a retentive memory, and a keen attention, and should be well read in the writings of eminent masters of the art. And even if he be well skilled in all these points, still he undertakes a subject full of hazard and difficulty, as the many errors into which the greatest men have fallen most abundantly evince. Beginners in criticism, therefore, should remember, that in such a subject the greatest caution must be exercised; especially with respect to the sacred books, which have been written out with much greater care than common works, in relation to which the same caution is nevertheless usually recommended. For this reason some general rules are added, which may lay down in a few words the mode of procedure adopted by the best critics.

1) The critic, like the judge, must be unbiassed by any partiality; in other words, he must be free from any preconceived opinion. It is therefore his duty, not only to be free from any desire to support or to alter a received reading, but also to abstain from giving any decision previously to a thorough examination of witnesses and of internal arguments.2) As the judge not only hears and examines the arguments of the plaintiff, but also those of the accused, so must the critic examine with equal attention, not only the arguments alleged against a received reading, but those also which are in its favour. In doing this he is to take care not to suppose that all the witnesses not produced in favour of a various reading are on the side of the received; for it often happens that many of them have no testimony to give, that is, that the manuscripts are defective in that place, or that it is impossible to infer from the versions what the translators read.3) As spurious readings have been very extensively propagated by means of the Masora, or of standard manuscripts from which many others have been written and corrected, or even on account of their being more orthodox; not every reading for which the greatest number of witnesses depose is to be considered as genuine. The true

reading may be preserved by a few, or by even a single one, as is the case with hy in Ps. xxxvii. 28. Witnesses, therefore, as in civil and criminal courts, are not only to be numbered but also scrutinized, their testimony examined what can be offered on the other side investigated, and internal arguments also carefully considered.

one.

4) If in all other respects two readings are equal, (a case which rarely occurs,) the number of witnesses must decide the question of genuineness, the probability being the greater in proportion as this increases. But from the number of witnesses are to be excluded, not only all the mediate versions, but also manuscripts of the same recension or family, which ought to be regarded as constituting only This, however, cannot take place until the manuscripts are reduced into families.-5) If much more ancient witnesses, or superior manuscripts, or stronger internal arguments, declare in favour of one reading, all things are by no means equal; for the more ancient witnesses are the more weighty, the oldest versions are of greater importance than manuscripts, the more ancient or excellent manuscripts are preferable to the more modern and inferior, and internal arguments very often add great weight to witnesses.-6) Readings which have the unanimous testimony of all or almost all the versions, manuscripts, and editions, and are favoured also by internal arguments, are critically certain. Since this is the case with all the readings of the Old Testament, with the exception of a few various readings, affecting the sense; it clearly follows, that, as a whole, the text is critically certain. Thus the incorruptness of the books of the Old Testament is more fully confirmed.

§ 144. Critical conjecture.

Thus far I have treated of the correction of the text from the testimony of the ancients. It remains to be inquired whether it can be allowed in any case to correct the Hebrew text, independently of any ancient testimony, from internal arguments simply, or from critical conjecture. Should this be absolutely forbidden, it would be necessary to admit, that no genuine reading has perished under the hands of transcribers during the lapse of so many centuries, in the five last of which, indeed, there are many manuscript witnesses ex

tant, but in the whole period which intervened between the 10th century after and the 3d before Christ none are to be found except some versions and certain quotations made by writers of those ages, and from the latter date up to the authors themselves no witnesses of readings exist. But if it is necessary to concede that genuine readings have perished, then it must also be acknowledged, that it is allowable by the aid of internal arguments or critical conjecture, to restore those spurious readings, at least, which change the meaning.

If criticism which conducts its operations on the authority of witnesses is to be carefully and moderately exercised, much rather ought this to be the case when independently of such authority it is driven to conjecture. To seek after conjectures designedly, or on light grounds to alter the reading, and introduce the alteration into the text, is what sound critics stigmatize as a censurable eagerness for correcting; and they lay down the principle, that conjectures are not to be attempted except when necessity forces, or very great probability persuades to such a course, and that even then they are not to be introduced into the text, but thrown into notes.

Necessity forces, when a learned and skilful interpreter, (for those who are not versed in the subject have nothing to do with a matter of this importance,) finds it impossible to elicit a suitable sense from all the readings of a place which are extant, after all other methods have been tried; or when the sense is repugnant to undoubted history well known to the author of the book, or to doctrine elsewhere proposed by the same writer. And in this case conjecture ought not to be sought, but freely to offer itself and to consist in a slight alteration of a word made in a way similar to that in which spurious readings arise. Thus it would seem that in Ps. cvii. 3., DD ought to be

.מימין changed into

.T.

Probability is much in favour of conjecture when a certain reading gives a sense, tolerable indeed, yet not in all respects adapted to the context, while a slight alteration corresponding with the manner in which errors arise produces a complete consistency. Thus in Jud. xiv. 14, 15., where it is said, "they could not ex

pound the riddle in three days, and it came to pass on the seventh day, &c.," if new or ', three, be changed into new or ', six,

every thing will be clear. If, however, the probability of a reading being erroneous is rather small, or if the alteration of the reading is considerable, or varies much from the analogy by which errors originate, the conjectural emendation is to be rejected, although it may suggest itself. Thus John David Michaelis conjectures without sufficient reason, that in I Sam. viii. 16, for in your asses, the

-:

reading ought to be in your precious things; but if a better

reading were wanted in this place, it is offered by the Alexandrine translator, who gives the meaning of Dopɔ, your oxen.

That sort of conjecture which aims at making the text conformable to grammatical rules is to be attributed solely to an unwarrantable eagerness for correcting.

§ 146. Doctrinal conjecture.

..:

An alteration of a reading without a witness on doctrinal grounds merely, is called doctrinal conjecture. This is nothing but a corruption of the text, and not to be tolerated in any book. It is sufficient to remark that such conjecture is founded not only on strictly doctrinal arguments, or such as refer to articles of faith or moral precepts, but also on all others derived from opinions previously fixed. If, therefore, doctrinal conjecture induced the Jews to alter p Gerizim into by Ebal in Deut. xxvii. 4, in order to deprive the Samaritans of an argument for that as a place of worship; it was the same principle which led the Samaritans to add in Exod. xii. 10. to the words "the sojourning of the Israelites in Egypt," the clause" and in Canaan;" because, through a preconceived opinion they conjectured that 430 years would not agree with four generations.-The critic must also be on his guard, not to be induced by doctrinal arguments to ascribe the more weight to witnesses or to internal arguments, because the reading for which they give evidence is favourable to certain opinions.