Keri and Kethib

Keri And Kethib (קרי וכתיב, plural וכתיבן קריין), so frequently found in the margins and footnotes of the Hebrew Bibles, exhibit the most ancient various readings, and constitute the most important portion of the critico- exegetical apparatus bequeathed to us by the Jews of olden times. On this subject we substantially adopt Ginsburg's article in Kitto's Cyclopcedia, s.v. SEE MASORAH.

I. Signification, Classification, and Mode of Indication of the Keri and Kethib.-The word קרַי, keri', may be either the imperative or the participle passive of the Chaldee verb קרָא, to call out, to read, and hence may signify "Read," or "It is read," i.e. the word in question is to be substituted for that in the text. כּתַיב, kethib', is the participle passive of the Chaldee verb כּתִב, to write, and signifies "It is written," i.e. the word in question is in the text. Those who prefer taking the word קרַי as participle, do so on the ground that it is more consonant with its companion כּתַיב, which is the participle passive. The two terms thus correspond substantially to the modern ones margin (Keri) and text (Kethib). We may add that the Rabbins also call the Keri מַקרָא, mikra', scripture, and the Kethib מָסוֹרָה, masorah', tradition; but, according to our ideas, these terms should be reversed.

The different readings exhibited in the Keri and Kethib may be divided into three general classes:

a. Words to be read differently from what they are written, arising from the omission, insertion, exchanging, or transposition of a single letter (כּתַיב וּקרַי קרַי וּכתַיב);

b. Words to be read, but that are not written in the text (קרַי ולֹא כתַיב); and,

c. Words written in the text, but that are not to be read (בּתַיב ולֹא קרַי).

a. The first general class (variations) comprises the bulk of the various readings, and consists of

1. Corrections of errors arising from mistaking homonyms, e.g. לא, the negative particle, for the similarly sounding לו, the pronoun, of which we have fifteen instances (comp. Ex 21:8; Le 11:21; Le 25:30; 1Sa 2:3; 2Sa 16:18; 2Ki 8:10; Ezr 4:2; Job 13:15; Job 41:4; Ps 100:3; Ps 139:16; Pr 19:7; Pr 26:2; Isa 9:2; Isa 63:9), and two instances in which the reverse is the case (1Sa 2:16; 1Sa 20:2). Besides noticing them in their respective places, the Masorah also enumerates them all on Le 11:15. The Talmud (Sopherim, vi) gives three additional ones, viz., 1Ch 11:21; Job 6:21; Isa 49:5. על for אל, of which we have four instances (1Sa 10:24; 1Ki 1:33; Job 7:1; Isa 65:7; Eze 9:5).

2. Errors arising from mistaking the letters which resemble each other, e.g. ב for כ (comp. Pr 21:29); ג for ז (Eze 25:7); ד for (1Sa 4:13); ד for ר, of which the Masorah on Pr 19:19, and Jer 21:14, gives four instances (2Sa 13:37; 2Ki 16:6; Jer 21:14; Pr 19:19); ה for ת (Jer 28:1; Jer 32:1); ה for ֹם (2Sa 23:13); ִח for ה, of which the Masorah on Pr 20:21 gives four instances (2Sa 13:37; Pr 20:21; Song 1:17; Da 9:24); ט for ש (1Sa 14:32); י for ו in innumerable instances; כ for ב in eleven cases (Jos 4:18; Jos 6:5,15; 1Sa 11:6,9; 2Sa 5:24; 2Ki 3:24; Ezr 8:14; Ne 3:20; Es 3:4; Job 21:13; ֹם for ה(Isa 30:32); צ for ָע (2Ki 20:4); ר for ד twice (Jer 2:20; Ezr 8:14); ת for ִח (Ec 12:6); ת for ה (2Ki 24:14; 2Ki 25:17; Jer 52:21).

3. Errors arising from exchanging letters which be. long to the same organs of speech, e.g. ב for מ of which the Keri exhibits one instance (Jos 22:7), and vice-versa, of which the Great Masorah, under letter ב, gives six instances (Jos 3:16; Jos 24:15; 2Ki 5:12; 2Ki 12:10; 2Ki 23:33; Da 11:18); ִח for א (2Ki 17:21); ָע for א (1Sa 20:24; 1Ki 1:33; Job 7:1; Isa 65:7; Eze 9:5); מ for פ (Isa 65:4).

4. Errors arising from the transposition of letters, which the Masorah designates מוקדם ומאוחר, and of which it gives sixty-two cases, as, for instance, the textual reading, or Kethib, is האהל, the tent, and the marginal reading, or Keri, transposing the letters ל and ה, has האלה, these (comp. Jos 6:13; Jos 20:8; Jos 21:27; Jg 16:26; 1Sa 14:27; 1Sa 19:18,22-23 [twice]; 27:8; 2Sa 3:25; 2Sa 14:30; 2Sa 17:16; 2Sa 18:8; 2Sa 20:14; 2Sa 24:16; 1Ki 7:45; 2Ki 11:2; 2Ki 14:6; 1Ch 1:46; 1Ch 3:24; 1Ch 27:29; 2Ch 17:8; 2Ch 29:8; Ezr 2:46; Ezr 4:4; Ezr 8:17; Ne 4:7; Ne 12:14; Es 1:5,16; Job 26:12; Ps 73:2; Ps 139:6; Ps 145:6; Pr 1:27; Pr 13:20; Pr 19:16; Pr 23:5,26; Pr 31:27; Ec 9:4; Isa 37:30; Jer 2:25; Jer 8:6; Jer 9:7; Jer 15:4; Jer 17:23; Jer 24:9; Jer 29:18,23; Jer 32:23; Jer 42:20; Jer 1; Jer 15; Eze 36:14; Eze 40:15; Eze 42:16; Eze 43:15-16; Da 4:9; Da 5:7,16 [twice], 29).

5. Errors arising from the small letter י being dropped before the pronominal ו from plural nouns, and making them to be singular, of which there are a hundred and thirteen instances [it is very strange that the Masorah Magna only enumerates fifty-six of these instances] (Ge 33:4; Ex 27:11; Ex 28:28; Ex 32:19; Ex 39:4,33; Le 9:22; Le 16:21; Nu 12:3; De 2:33; De 7:9; De 8:2; De 27:10; De 33:9; Jos 3:4; Jos 8:11; Jos 16:3; Ru 3:14; 1Sa 2:9-10 [twice]; 3:18; 8:3; 10:21; 22:13; 23:5; 26:7 [twice], 11, 16; 29:5 [twice]; 30:6; 2Sa 1:11; 2Sa 2:23; 2Sa 3:12; 2Sa 12:9,20; 2Sa 13:34; 2Sa 16:8; 2Sa 18:7,18; 2Sa 19:19; 2Sa 20:8; 2Sa 23:9,11; 2Sa 24:14,22; 1Ki 5:17; 1Ki 10:5; 1Ki 18:42; 2Ki 4:34; 2Ki 5:9; 2Ki 11:18; Ezr 4:7; Job 9:13; Job 14:5; Job 15:15; Job 20:11; Job 21:20; Job 24:1; Job 26:14; Job 31:20; Job 37:12; Job 38:41; Job 39:26,30; Job 40:17; Ps 10:5; Ps 24:6; Ps 58:8; Ps 106:45; Ps 147:19; Ps 148:2; Pr 6:13 [twice]; 22:24; 26:24; Isa 52:5; Isa 56:10; Jer 15:8; Jer 17:10-11; Jer 22:4; Jer 32:4; Jer 52:33; La 3:22,32,39; Eze 3:20; Eze 17:21; Eze 18:23-24; Eze 31:5; Eze 33:13,16; Eze 37:16 [twice], 19; 40:6, 22 [twice], 26; 43:11 [thrice], 26; 44:5; 47:11; Da 11:10; Am 9:6; Ob 1:11; Hab 3:14); as well as from the insertion of י before the pronominal ו and before the pronominal in singular nouns, and making them plural; the Keri exhibits seven instances of the former.(1Ki 16:26; Ps 105:18,28; Pr 16:27; Pr 21:29; Ec 4:16; Da 9:12) and eight of the latter in the word דבוֹ (Jg 13:17; 1Ki 8:26; 1Ki 22:13; Ps 119:147,161; Jer 15:16 [twice]; Ezr 10:12).

6. Errors of a grammatical nature, arising from dropping the article הwhere it ought to be, of which the Keri exhibits fourteen instances (1Sa 14:32; 2Sa 23:9; 1Ki 4:7; 1Ki 7:20; 1Ki 15:18; 2Ki 11:20; 2Ki 15:25; Isa 32:15; Jer 10:13; Jer 17:19; Jer 40:3; Jer 52:32; La 1:18; Eze 18:20), or from the insertion of it where it ought not to be, of which there are ten instances (1Sa 26:12; 1Ki 21:8; 2Ki 7:12-13; 2Ki 15:25; Ec 6:10; Ec 10:3,20; Isa 29:11; Jer 38:11); or from the dropping of the הafter נער, or writing הוא, instead of היא, when used as feminine.

7. Errors arising from the wrong division of words, e.g. the first word having a letter which belongs to the second, exhibited by the Keri in three instances, and stated in the Masorah on 2Sa 5:2' (2Sa 5:2; Job 38:12; La 4:16), or the second word having a letter which belongs to the first, of which there are two instances (1Sa 21:12; Ezr 4:12); or one word being divided into two separate words, of which the Masorah on 2 Chronicles 34 mentions eight instances (Jg 16:25; 1Sa 9:1; 1Sa 24:8; 1Ki 18:5; 2Ch 34:6; Isa 9:6; La 1:6; La 4:3), or two separate words being written as one, exhibited by the Ker. in fifteen instances (Ge 30:11; Ex 4:2; De 33:2; 1Ch 9:4; 1Ch 27:12; Ne 2:20; Job 38:1; Job 40:6; Ps 10:10; Ps 55:16; Ps 123:4; Isa 3:15; Jer 6:29; Jer 18:3; Eze 8:6).

8. Exegetical Keris or marginal readings which substitute euphemisms for the cacophonous terms used in the text, in accordance with the injunction of the ancient sages, that " all the verses wherein indecent expressions occur are to be replaced by decent words (e.g. ישגלנה by ישכבנה [of which the Keri exhibits four instances, viz. De 28:30; Isa 13:16; Jer 3:2; Zec 14:2]; עפולים by טחורים [of which the Keri exhibits six instances, viz. De 28:27; 1Sa 5:6,9; 1Sa 6:4-5,17; omitting, however, 1Sa 5:12]; חריונים by דביונים [of which the Keri exhibits one instance, viz. 2Ki 6:25]; חוריהם by צואתם [of which the Keri exhibits two instances, 2Ki 18:27; Isa 36:12]; מימי שיניהם by מימי רגליהם [of which the Keri exhibits two instances, 2Ki 18:27; Isa 36:12]; למחראות by למוצאות [of which there is one instance, 2Ki 10:27, comp. Megilla, 25 b])." The manner in which this general class of various readings is indicated is as follows: The variations specified under 1 and 2, not affecting the vowel points, are simply indicated by a small circle or asterisk placed over the word in the text (כתיב), which directs to the marginal reading (קרי), where the emendation is given, as, for instance, the Kethib in Ex 21:8 is לֹא, in 1Sa 20:24 עֶל, and in Pr 21:29 יַָֹבין, and the marginal gloss remarks ק יבין ק8 אל ק8 לו, the ק being an abbreviation for קרי. In the variations specified under 3 and 4, where the different letters of the Kethib and the Keri require different vowel points, the abnormal textual reading, or the Kethib, has not only the small circle or asterisk, but also takes the vowel points which belong to the normal marginal reading, or the Keri, e.g. the' appropriate pointing of the textual reading, or the Kethib, in 2Ki 17:21, is וִיֶּדֶא, but it is pointed וִיִּדִּאֹ, because these vowel signs belong to the marginal reading, or the Keri, וידח, which it is intended should accompany the vowel points in the text. The same is the case with the textual reading in 2Sa 14:30, which, according to the marginal reading, exhibits a transposition of letters, and which can hardly be pronounced with its textual points והִוצַּיתיֹּה, because these vowel signs belong to the Keri, והציתוה. Finally, in the variations specified under 5, 6, 7, and 8, which involve an addition or diminution of letters, and which have therefore either more or fewer letters than are required by the vowel points of the Keri, a vowel sign is sometimes given without any letter at all, or two vowel signs have.to be attached to one letter, and sometimes a letter has to be without any vowel sign; the variation itself being either indicated in the margin by the exhibition of the entire word which constitutes the differerent reading, or by the simple remark that such and such a letter is wanting or is redundant. For instance, in La 5:7, which, according to the Masorah, exhibits two of the twelve instances where the ו conjunctive has been dropped from the beginning of words (comp. also 2Ki 4:7; Job 2:7; Pr 23:24; Pr 27:24; Isa 55:13; La 2:2; La 4:16; La 5:3,5; Da 2:43), the textual reading, or Kethib, is ִאֲנַחַנוּ אֵינָם and the marginal reading, or Keri, is ואנחנו ואינם, the vowel sign of the conjunction from the margin being inserted in the text under the little circle, which, consequently, has no letter at all; in Jer 42:6, again, where the textual reading is אנו, and the marginal reading אנחנו, yet the Kethib, which has only three letters, takes the vowel signs of the Keri, which has five letters, and is pointed אֲנִוּ, with. two different vowel points attached to the one ו; whilst in 2Ki 7:15, where the reverse is the case, the marginal reading having fewer letters, and hence fewer vowels than the textual reading, which takes the vowel signs of the former, the Kethib is pointed בּהֹחָפזָם; and the ה has no vowel sign at all. There is a peculiarity connected with the marginal indication of those words the variations of which consist in the diminution or addition of a single letter. When a letter is dropped from a word in the text, the whole word is given in the marginal reading with the letter ii' question, and the remark "Read so;" as, for instance, 1Sa 14:32; Pr 23:24, where the ה, according to the Masorah, is dropped from השלל, and ו from ויולד, as indicated by שּׁלָל and יֹוֹלֵג; the marginal glosses are ק8 ויולד ק8 השלל; but when the reverse is the case, if a letter has crept into a word, the whole word is not given in the marginal gloss, but it is simply remarked that such and such a letter is redundant (יתיר), or is not to be read (לא קרי), as, for instance, in Ec 10:20; Ne 9:17, where the ה, according to the Masorah, has crept in before כנפים, and ו before חסד, the marginal gloss simply remarks יתיר ו8 יתיר ה8. Upon this point, however, the greatest inconsistency is manifested in the Masoretic glosses; compare, for instance, the Kethib עיניו and ִרגלי in Ec 4:8,16, both of which, according to the Keri, have a redundant י, and are singular nouns, vet the Masoretic note upon the former is ק8 עינו, exhibiting the whole word, whilst on the latter it simply remarks יתיר י.

b. The second class (insertions directed), which comprises entire words that have been omitted from' the text, exhibits ten such instances which occur in the Hebrew Bible, as follows: Jg 20:13; Ru 3:5,17; 2Sa 8:3; 2Sa 16:23; 2Sa 18:20; 2Ki 19:31,37; Jer 31:38; Jer 1; Jer 29. Besides being noted in the marginal glosses on the respective passages, these omissions are also given in the Masorah ,on Deuteronomy 1 and Ru 3:16. They are also enumerated in the Talmud (Tract Sopherim, 6:8, and in Nedarim, 37 b). In Nedarim, however, the passage which refers to this subject is as follows: "The insertion of words in the text (ולא כתיבן קריין) is exhibited in פרת [2Sa 8:3]; איש [ibid. 16:23]; באים [Jer 31:38]; לה [ [Jer 50:29]; את[Ru 2:11]; אלי [Ru 3:5,17] ;" thus :omitting four instances, viz. Jg 20:13; 2Sa 18:20; 2Ki 19:31,37, and adding one, viz., Ruth ii, 11, which is neither given by the Masorah nor in Sopherim.

This class of variations is indicated by a small circle or asterisk placed in the text with the vowel signs of the word which is wanting, referring to the margin, where the word in question is given. Thus, for instance, in Jg 20:13, where, according to the Keri, the word בנֵי is omitted, the Kethib is בַניָמַן ° ולַֹא אָבוּ, upon which the marginal gloss remarks כתיב בני קרי ולאc. Of the third class (omissions suggested), exhibiting entire words which have crept into the text, there are eightinstances, as follows: Ru 3:12; 2Sa 13:33; 2Sa 15:21; 2Ki 5:18; Jer 38:16; Jer 39:12;

51:3; Eze 48:16. These variations are not only noted in the marginal glosses on the respective passages, but are also given in the Masorah on Ru 3:12. The passage in Nedarim, 27 b, which speaks of this class of variations, remarking, "Words which are found in the text, but are not read (כתיבן ולא קריין), are exhibited in נא [2Ki 5:18]; ואת [Jer 32:11]; ִידר [Jer 51:3]; חמש [Eze 48:16]; אם [Ru 3:12]," omits 2Sa 13:33; 2Sa 15:21; and Jer 38:16; Jer 39:12; and adds Jer 32:11, which dces not exist in the Masorah; whilst Sopherim, 6:9, which remarks אמנון כאשר במקום גואל ידר ִחמש, referring to 2Sa 13:33; Jer 39:12; 2Sa 15:21; Ru 3:12; Jer 51:3; Eze 48:16; omits 2Ki 5:18, and Jer 38:16.

This class of variations is not uniformly indicated in the different editions of the Bible. Generally the word in question has no vowel signs, but an asterisk or small circle is put over it, referring to the margin, where it is simply remarked כתיב ולא קרי, written [in the text], but not [to'be] read; in one or two instances, however, the word itself is repeated in the margin, as in 2 Kings, 5:18, where we have it נא כתיב ולא קרי, [the word] נא [is] written [in the text], but [is] not [to be] read.

II. Number and Position of the Keri and Kethib.-A great difference of opinion prevails about the number and position of these various readings. The Talmud, as we have shown above, and the early commentators, mention variations which do not exist in the Keris and Kethibs of the Masorah. This, however, is beyond the aim of the present article, which is to investigate the Keri and Kethib as exhibited in the Masorah and in the editions of the Hebrew Bible. From a careful perusal and collation of the Masorah, as printed in the Rabbinic Bibles, we find the following to be the number of the Keris and Kethibs in each book, according to the order of the Hebrew Bible:

Ge 24; Hab 2; Ex 12; Zep 1; Le 5; Hag 1; Nu 11; Zec 7; De 24; Mal 1 Joshua 38 Psalms 74 Judges 22 Proverbs 70

1 Samuel 73 Job 54 2 Samuel 99 Song of Songs 5 1 Kings 49 Ruth 13 2 Kings 80 Lamentations 28 Isa 5; Ec 11 Jeremiah 148 Esther 14 Ezekiel 143 Daniel 129 Hosea 6 Ezra 33 Joel 1 Nehemiah 28 Amos 3 1 Chronicles 41 Ob 1:1 2 Chronicles 39 Micah 4 Nahum 4 Total........... 1353

The disparity between Abrabanel's calculations about the number, of Keris and Kethibs, leading him to the conclusion that the Pentateuch has 65, Jeremiah 81, and 1 and 2 Samuel 138' (Introduction to Jeremiah), and the numbers which we have stated as existing in these books, is easily accounted for when it is remembered that this erudite commentator died fifteen years before the laborious Jacob b.-Chajim collated and published the Masorahs on the Hebrew Scriptures, and therefore had no opportunity of consulting them carefully. But we find it far morea difficult to account for the serious difference in the calculations of later writers and'our results, as may be seen from the table on the following page.

For the collation of Bomberg's Bible, the Plantin Bible, and the Antwerp Bible, we are indebted to the tables exhibited in Cappellus's Critica Sacra, p. 70, and Walton's Prolegomena (ed. Cantabrigiae, 1828, i, 473); and though we have been able by our arrangement to correct their blunder in representing Elias Levita as separating the Five Megilloth from the Hagiographa, and giving the number of Keris to be 329 exclusive of the Megilloth, yet we were, obliged to describe the Megilloth apart from the Hagiographa. to which they belong according to the Jewish order of the Canon. Elias Levita's own words on the numbers are as follows: "I counted the Keris and Kethibs several times, and found that they were in all 848.; of these, 65 are in the Pentateuch, 454 in the Prophets, and 329 in the Hagiographa. It is surprising that there should only be 65 in the Pentateuch, 22 of which refer to the single word נערה, which "interpretations," יתיר ; "Deficiencies," חסיר is נעי il the Kethib, and

נערה in the Keri; that the book of Joshua, which in quantity is about a tenth part of the Pentateuch, should have 32; and that the books of Samuel, which are merely about a fourth the size of the Pentateuch, 'should contain 133" (Massoreth HaMassoreth, ed. Sulzbach, 1771, p. 8 sq.). It will be seen from this extract that Elias Levita not only gives six Keris less in Joshua than we have given, but also differs from Abrabanel in the number of Keris to be found in the books of Samuel.

Bloomberg 1523-24 Plantin Bible 1566 Antwerp Royal Bible 1572 Elias Levita Our Results Pentateuch Variations 73 Interpolations Deficiencies 1

74 74 69 1 1 2 1 77 71 65 76

Earlier Prophets

Later Prophets Variations Interpolations Deficiencies

Variations Interpolation s Deficiencies 337 11 2 350 348 2

350 239 25 5 269 250 25 1 276 277 18 5 300 347 11

358 361

454 377

Five Megiloth Variations 51 Interpolations Deficiencies 11 62 43 48 71 14 8 57 56

Hagiographa Variations Interpolations Deficiencies 362 60 1 423 187 34 1 222 242 20 1 263 329 468 Total 1259 901 1048 848 1353

N.B. - In this table, what are denoted by "Variations" are designated by the Marosites as קרי ;

III. Origin and Date of the Keri and Kethib.-The Talmud traces the source of these variations to Moses himself, for we are distinctly told in Nedarim, 37 b, that "the pronunciation of certain words according to the scribes (מקרא סופרים), the emendations of the scribes (עטור סופרים), the not reading of words which are in the text (כתיב ולא קרי), and the reading of words which are not in the text (קרי ולא כתיב), etc., are a law of Moses from Sinai." Jacob b.-Chajim defends this view in his elaborate Introduction to the Rabbinic Bible. Elias Levita, who also expresses this Talmudic declaration, explains it as follows: " The Keri and Kethib of the Pentateuch only are a law of Moses from Mount Sinai, and the members of the Great Synagogue, Haggai, Zechariah, Malachi, Daniel, Hananiah, Mishael, Azariah, Ezra, Nehemiah, Mordecai, and Zerubbabel, and other wise men from the craftsmen and artisans (מהחרש והמסגר) to the number of a hundred and twenty, wrote down the Keri and the Kethib according to the tradition which they possessed that our teacher Moses (peace be with him!) read words differently from what they were written in the text; this being one of those mysteries which they knew, for Moses transmitted this mystery to Joshua, Joshua to the elders, the elders to the prophets, etc., and these were put down in the margin as his readings, Ezra acting as a scribe. In the same manner they proceeded in the Prophets and Hagiographa with every word respecting which they had a tradition orally transmitted from the prophets and the sages that it was read differently from what it was in the text. But they required no tradition for the postexilian books, as the authors themselves were present with them; hence, whenever they met with a word which did not seem to harmonize with the context and the sense, the author stated to them- the reason why he used such anomalous expressions, and they wrote down the word in.the margin as it should be read" (Massoreth Ila-Massoreth, fol. 8 b, sq.). Mendelssohn, in his valuable introduction to his translation of the Pentateuch, and most of the ancient Jewish writers, propounded the same view. It is in accordance with this recondite sense ascribed to the origin of. the Keri and Kethib that Rashi remarks on Ge 8:16, "The Keri is הוצב, the Kethib היצא, because he was first to tell them to go out; but if they should refuse to go, he was to make them go." Kimchi, however, is of the opposite opinion. So far from believing that these variations proceeded from the sacred writers themselves, who designed to convey thereby various mysteries, he maintains that the Keri and Kethib originated after the Babylonian captivity, when the sacred books were collected by the members of the Great Synagogue. These editors of the long-lost and mutilated inspired writings "found different readings in the volumes, and adopted th'ose which the majority of copies had, because these, according to their opinion, exhibited the true readings. In some places they wrote down one word in the text without putting the vowel signs to it, or noted it in the margin without inserting it in the text, whilst in other places they inserted one reading in the margin and another in the text" (Introduction to his Commentary on Joshua). Ephodi (flourished 1391-1403), who maintains the same view, remarks that Ezra and his followers "made the Keri and Kethib on every passage in which they found some obliterations and confusion, as they were not sure what the precise reading was." Abrabanel, who will neither admit that the Keris and Kethibs proceeded from the sacred writers themselves, nor that they took their rise from the imperfect state of the codices, propounds a new theory. According to him, Ezra and his followers, who undertook the editing of the Scriptures, found the sacred books entire and perfect; but in perusing them these editors discovered that they contained irregular expressions, and loose and ungrammatical phrases, arising from the carelessness and ignorance of the inspired writers. " Ezra had therefore to explain these words in harmony with the connection, and this is the origin of the Keri which is found in the margin of the Bible, as this holy scribe feared to touch 'the words which were spoken or written by the Holy Ghost. These remarks he made on his own account to explain those anomalous letters and expressions, and he put them in the margin to indicate that the gloss is his own. Now, if you examine the numerous Keris and Kethibs in Jeremiah, and look into their connection, you will find them all to be of this nature, viz., that they are to be traced to Jeremiah's careless and blundering writing. .... From this you may learn that the books which have most Keris and Kethibs show that their authors did not know how to speak correctly or to write properly" (Introduction to his Commentary on Jeremiah). Though Abrabanel's hypothesis has more truth in it than the other theories, yet it is only by a combination of the three views that the origin of the Keri and Kethib can be traced and explained. For there can be no doubt that some of the variations, as the Talmud, Rashi, etc., declare, have been transmitted by tradition from time immemorial, and have their origin in some recondite meaning or mysteries attached to the passages in question; that some, again, as Kimchi, Ephodi, etc., rightly maintain, are due to' the blunders and corruptions which have crept into the text in the course of time, and which the spiritual guides of the nation tried to rectify by a comparison of codices, as is also admitted by the. Talmud (comp. Jerusalem Megillah, 4:2; Sopherim, 6:4); and that others, again, as Abrabanel remarks, are owing to the carelessness of style, ignorance of idioms and provincialisms, which the editors and successive interpreters of the Hebrew canon discovered in the different books, or, more properly speaking, which were at variance with the grammatical rules and exegetical laws developed in aftertime by the Masorites. Such, however, was their reverence for the ancient text, that these Masorites who made the new additions to it left the text itself untouched in the very places where they believed it necessary to follow another explanation or reading, but simply inserted the emendation in the margin. Hence the distinction between the ancient text as it was written, or Kethib (כתיב), and the more „modern emended reading, or Keri (קרי); and hence, also, the fact that the Keri is not inserted in the synagogal scrolls, though it is followed in the public reading of the Scriptures.

IV. Importance of the Keri and Kethib, especially as relating to the English Version of the Hebrew Scriptures.-Some idea of the importance of the Keri and Kethibmay be gathered from the following analysis of the seventy-six variations which occur in the Pentateuch. Of the seventy-six Keris, twenty-one give נערה instead of נער (Ge 24:14,16,28,55,57; Ge 34:3 [twice], 12; De 22:15 [twice], 16, 20, 21, 23, 24, 25, 26 [twice], 27, 28, 29), which was evidently epicene in earlier periods (comp. Gesenius, G-qamm. sec. 23, sec. 32, 6; Ewald, Lehrbuch, sec. 175, b); fifteen have the plural termination אָּיו affixed to nouns instead of the singular ו in the text (Ge 33:4; Ex 27:11; Ex 28:28; Ex 32:19; Ex 39:4,33; Le 9:22; Le 16:21; Nu 12:3; De 2:33; De 5:10; De 7:9; De 8:2; De 27:10; De 33:9), which some think is no real variation, since in earlier periods the termination ו was both singular and plural, just as בגדי stands for both בַּגדַּי. and בּגָדֵי; seventeen give more current and uniform forms of words (Ge 8:17; Ge 10:19; Ge 14:8; Ge 24:33 with 50:26; 25:23 with 35:11; 27:3 with 5, 7; 27:29 with the same word in the next clause; 36:6,14 with ver. 18; 39:20, 22; 43:28 with 27:29; Ex 16:2,7 with Nu 16:11; Nu 14:36 with 15:24; Nu 21:32 with 32:39; 32:7 with 30:6; De 32:13 with Am 4:13); five substitute the termination third person singular, ו for ה (Ge 49:11 [twice]; Ex 22:26; Ex 32:17; Nu 10:36), which is a less common pronominal suffix (comp. Gesenius, Gramm. sec. 91; Ewald, Lehrbuch, sec. 247, a); two make two words of one (Ge 30:11; Ex 4:2); two have שליו instead of שלו (Ex 16:13; Nu 11:32); three give plural verbs instead of singular (Le 21:5; Nu 34:4; De 31:7), which are no doubt an improvement, since Nu 34:4 is evidently a mistake, as may be seen from a comparison of this verse with verse 5; three substitute the relative pronoun כֹו for the negative particle לא (Ex 21:8; Le 11:21; Le 25:30), which is very important; two substitute euphemisms for cacophonous expressions (De 28:27,30); and two are purely traditional, viz., Nu 1:16; Nu 26:9. The Pentateuch, however, can hardly be regarded as giving an adequate idea of the importance of the Keri and Kethib, inasmuch as the Jews, regarding the law as more sacred than any other inspired book, guarded it against being corrupted with greater vigilance than the rest of the canon. Hence the comparatively few and unimportant Keris when contrasted with those occurring in the other volumes. Still, the Pentateuch contains a few specimens of almost all the different Keris.

As to the question how far our English versions have been influenced by the Keri and Kethib, this will best be answered by a comparison of the translations with the more striking variations which occur in the Prophets and Hagiographa. In Jos 5:1, the textual reading is "till we were passed over" עברנו), the Keri has עברם, " until they passed over;" and though the Sept., Vulg., Chaldee, Luther, the Zurich Bible, Coverdale, the Bishops' Bible, the Geneva Version, etc., adopt the Keri, the A. V., following Kimchi, adheres to the Kethib; whilst in Jos 6:7, where the textual reading is "and they said (ויאמרו) unto the people," and the marginal emendation is "and he said" (ויאמר), and where the Vulg., Chaldee, Luther, the Zurich Bible, Coverdale, the Bishops' Bible, and the Geneva Version again adopt the Keri, as in the former instance, the A. V. abandons the textual reading and espouses the emendation. In Jos 15:47, where the Keri is "the bordering sea (הגבול הים) and its territory," and the Kethib has "and the great sea (הים הגדל) and the territory," which is again followed by the ancient versions and the translations of the Reformers, the A. V., without taking any notice of the textual reading in the margin, as in Jos 8:16, adopts the emendation, whereas in Jos 15:53 the A. V. follows the textual reading (ינום) Janum, noticing, however, the emendation (ינום) Janus'in the margin. All the ten emendations of the second class, which propose the insertion of entire words into the text (ולא כתיב קרי), are adopted in the A. V. without the slightest indication by the usual italics that they are not in the text. Of the eight omissions of entire words-in the third class (כתוב ולא קרי) nothing decisive can be said, inasmuch as six of them. refer to simple particles, and they might either be recognised by the translators or not without its being discernible in the version. The only two instances, however, where there can be no mistake (Jer 41:3; Ezekiel xlviii, 16), clearly show that the A. V. follows the marginal gloss, and accordingly rejects the words which are in the text. Had the limits of this article allowed it, we could have shown still more unquestionably that, though the A. V. generally adopts the marginal emendations, yet in many instances it proceeds most arbitrarily, and adheres to the textual reading; and that, with very few exceptions, it never indicates, by italics or in the margin, the difference between, the textual and the marginal readings.

Inattention to the Keri and Kethib has given rise to the most fanciful and absurd expositions, of which the following may serve both as a specimen and a warning. In looking at the text of the Hebrew Bible, it will be seen that there is a final Mem (ֹם) in the middle of the word לםרבה, Isa 9:6. We have already alluded to the fact that it exhibits one of the fifteen instances where the Kethib, or the textual reading, is one word, and the Keri, or the emended reading, proposes two words (see above, sec. 1). Accordingly, לםרבה stands for לָם רִבֵּה = לָהֶם, i.e. "to them the dominion shall be great," corresponding to the common abbreviation בָּם for בָּהֶם. The question is not whether לם may be considered as an abbreviation of להם, seeing there are no other examples of it; suffice it to say that Jewish scribes and critics of ancient times took it as such, just as they regarded אראלם (Isa 33:7) as a contraction of להם אראה = לם (comp. the Syriac, Chaldee, Aquila, Symmachus, Theodotion, Vulgate, Elias Levita, etc.); and that the Sept. read it as two words (i.e. לה רבה). Subsequent scribes, however, found it either to be more in accordance with the primitive reading, or with their exegetical rules, as well as with the usage of the prophet himself (comp. Isa 33:23), to read it as one word; but their extreme reverence for the text prevented them from making this alteration without indicating that some codices have two words. Hence, though they joined the two words together as one, they yet left the final Mem to exhibit the variation. An example of the reverse occurs in Ne 2:13, where המפרוצים has been divided into two words, המ פרוצים, and where the same anxiety faithfully to exhibit the ancient reading has made the editors of the Hebrew canon retain the medial Mem at the end of the word. It was to be expected that those Jews who regard both readings as emanating from the Holy Spirit, and as designed to convey some 'recondite meaning, would find some mysteries in this final Mem in the middle of לםרבה. ' Hence we find in the Talmud (Sanhedrin, 94) the following remark upon it: "Why is it that all the Mems in the middle of a word are open [i.e. מ] and this one is closed [i.e. ֹם] ? The Holy One (blessed be he!) wanted to make Hezekiah the Messiah, and Sennacherib Gog and Magog; whereupon Justice pleaded before the presence of the Holy One (blessed be he !),'Lord of the World, ' What! David the king of Israel, who sang so many hymns and praises before thee, wilt thou not make him the Messiah; but Hezekiah, for whom thou. hast performed all those miracles, and who has not uttered one song before thee, wilt thou make him the Messiah?' Therefore has the Mem been closed." Aben-Ezra again tells us that the scribes (not he himself, as Gill erroneously states) see in it an allusion to the recession of the shadow on the dial in Hezekiah's time; whilst Kimehi will have it that it refers to the "stopping up of the breaches in the walls of Jerusalem, which are broken down during the captivity, and that this will take place in the days of salvation, when the kingdom which had been shut ip till the coming of the Messiah will be opened." But that Christian expositors should excel these mystical interpretations is surpassing strange. What are we to say to Galatinus, who submits that this Mem, being the cipher of 600, intimates that six hundred years after this prophecy the birth of Christ was to take place? or to the opinion which he quotes, that the name שרה מרים Miaria Dominna, or even the perpetual virginity of Mary is thereby indicated (lib. 7:c. xiii)?. or to Calvin, who thinks that it denotes the close and secret way whereby the Messiah should come to reign and set up his kingdom? or to the opinion which he mentions that it indicates the exclusion of the Jews from the Messiah's kingdom for their unbelief? or to the conjecture of Gill, that "it may denote that the government of Christ, which would be for a time straitened, and kept in narrow bounds and limits, should hereafter be throughout the world, to the four corners of it, so as to be firm and stable, perfect and complete, which the figure of this letter, being shut and four- square, may be an emblem of?" It should be added that there are some words which are always read differently (קרי) from what they are written in the text (כתיב), and which, from the frequency of their occurrence, have only the vowel signs of the proposed Keri, without the latter being exhibited in the marginal gloss. These are,

a. The name יהוה, which has always the vowel signs of אֲדֹנָי, and is pronounced with these vowels, i.e. יהוֹה, except when it precedes this name itself, in which case it has the vowel signs of אֵֹּלהַים, i.e. יהוַה.

b. The name Jerusalem, when, as in the earlier books of Scripture, it is written with a Yod before the Mens, has never its own points, i.e. ירוּשָׁלֵם or אָּם, but has the vowel signs of ירוּשָׁלִיַם and is read so; c. The word הוּא, which was epicene in earlier periods, is always pointed הַוא in the Pentateuch, when it is used as feminine, to make it conformable to the later feminine form היא; and,

c. The name יששכר is always furnished with the vowels belonging to the Keri, יַשָּׂכָר with one Shin.

It remains only for us to say under this head that the judicious critic will often find good reason for differing from the opinion that seems to be implied in these Masoretic notes, and will in such cases, of course, prefer the Kethib to the Keri. SEE CRITICISM, BIBLICAL.

V. Literature.-One of the earliest attempts freely to discourse upon the origin and value of the Keri and Kethib is that of D. Kimchi, in the Introduction to his Commentary on Joshua; Abrabanel, too, has a lengthy disquisition on this subject, in the Introduction to his Commentary on Jeremiah. He was followed by the laborious Jacob ben-Chajim, who fully discusses the Keri and Kethib in his celebrated Introduction to' the Rabbinic Bible, translated by Ginsburg in the Journal of Sacred Literature for July, 1863; and by the erudite and bold Elias Levita, who gives a very lucid account of the Keri and Kethib in his Massoreth Ha-Massoreth, ed. Sulzbach, 1771, p. 8 a, sq.; 21 a, sq. Of Christian' writers are to be mentioned the masterly treatises by Cappellus, Critica Sacra, lib. 3:cap. 9:sq.; Buxtorf, Tiberias, cap. xiii; Buxtorf the younger, Anticritica (Basileae, 1653), cap. 4:p. 448-509; Hilleri De Arcano Kethib et Keri (Tub. 1692); Walton, Biblia Polyglotta, Proleg. (Cantab. 1828), i, 412 sq.;

Wolf, Bibliotheca Hebrcea, ii, 507-533; Frankel, Vorstudien-zu der Septuaginta (Leipzig, 1841), p. 219 sq.; Sticht, De Keri et Kethibh (Altouia, 1760; and against him Dreschler, Sententic Stichii, etc. Lips. 1763); Tragard, כתיב וקרי (Gryph. 1775);W.olffradt, De Keri et Ch'ihibh (Rost. 1739). SEE VARIOUS READINGS.