SV-. 
 
 a 
 
 .as 
 
 pq » J^-A 
 
 1 <y 
 P 
 
 : <U 
 : *- 
 
 T 
 
 : ill 
 
 *e 
 
 is? . 
 
 rrf — 
 4> — 
 
 Pbl 
 
 -V 
 
 ■ > 
 
 its 
 
 i P 
 +-> 
 
 0>. 
 
 (6 
 
 'J) 
 
 1 
 
 ill* 
 J- 
 
 ■|C0 
 
 o 
 
 iffl 
 
 Hh4 x$^£t 
 
 
 
 
 
 3*K 
 
 I '%fe J 
 
 
 *i- ufer 
 
 
 KE 
 
The date shows when tms voiurhe w^S IS886 
 
 To renew this book copy the call No. and give 
 to the librarian. 
 
 MAY 
 
 (tynrtttU Htmtteraittj Ki 
 
 HOME USE RULES 
 
 All Books subject to recall 
 
 All borrowers must regis- 
 ter in the library to bor- 
 row books for home use. 
 
 All books must be re- 
 turned at end of college 
 year for inspection and 
 repairs. 
 Limited books must be 
 returned within the four 
 week limit and not renewed. 
 
 Students must return all 
 books before leaving town. 
 Officers should arrange for 
 the return of books wanted 
 ■ during their absence 1 from 
 town. 
 
 Volumes 1 of periodicals 
 and of pamphlets are held 
 in the library as much as 
 possible. For special pur- 
 poses they are given out 
 for a limited time. 
 
 Borrowers should not use 
 their library privileges for 
 the benefit of other persons. 
 
 Books of special value *, 
 and gift books, when the 
 giver wishes it, are not al- 
 lowed to circulate. 
 
 Readers are asked to re- 
 port all cases of books 
 marked or mutilated. 
 
 Do not deface books by marks and writing. 
 
 Jiliara, Nm» Hork 
 
 Augus tana Library 
 
Cornell University Library 
 BS1322 .B74 
 + 
 Alternative readings in the Hebrew of th 
 
 3 1924 031 022 183 
 
 olin Overs 
 
'/M 
 
 Cornell University 
 Library 
 
 The original of this book is in 
 the Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924031022183 
 
AUGUSTANA . LIBRARY . PUBLICATIONS 
 
 NUMBER 8 
 
 ALTERNATIVE READINGS 
 
 IN '1 HE 
 
 HEBREW 
 
 OF THE 
 
 BOOKS OF SAMUEL 
 
 BY 
 
 OTTO H. BOSTROM, Ph. D. 
 
 PUBLISHED 
 
 BY THE AUTHORITY OF THE BOAKD OF DIEECTOES OF 
 
 AUGUSTANA COLLEGE AND THEOLOGICAL SEMINABY 
 
 BOCK ISLAND, ILLINOIS 
 
 ROCK ISLAND, ILL. 
 
 AUGUSTANA BOOK CONCEEN, PRINTERS 
 1918 
 
 Fa- 
 
AUGUSTANA LIBRARY PUBLICATIONS 
 
 No. 
 
 1. The Mechanical Composition of Wind Deposits, by Johan 
 August Udden. 
 
 2. An Old Indian Village, by Johan August Udden. 
 
 3. The Idyl in German Literature, by Gustav Albert Andreen. 
 
 4. On the Cyclonic Distribution of Rainfall, by Johan August 
 Udden. 
 
 5. Fossil Mastodon and Mammoth Remains in Illinois and Iowa, 
 by Netta C. Anderson, 
 
 and 
 Proboscidean Fossils of the Pleistocene Deposits in Illinois 
 and Iowa, by Johan August Udden. 
 6.. Scandinavians Who Have Contributed to the Knowledge of 
 the Flora of North America, by Per Axel Rydberg, Ph. D., 
 
 and 
 A Geological Survey of the Lands Belonging to the New York 
 and Texas Land Company, Ltd., in the Upper Rio Grande 
 Embayment in Texas, by Johan August Udden. 
 7. Genesis and Development of Sand Formations on Marine 
 Coasts, by Pehr Olsson-Seffer, Ph. D., 
 
 and 
 The Sand Strand Flora of Marine Coasts, by Pehr Olsson- 
 Seffer, Ph. D. 
 
J 
 
 d 
 2 
 
 p 
 
 m 
 
 W 
 o 
 
 o 
 
 tri 
 
 H 
 O 
 H 
 J 
 
 J 
 O 
 
 o 
 
 <n 
 z 
 < 
 
 H 
 cc 
 P 
 CS 
 
 P 
 < 
 
 < 
 a 
 m 
 
 < 
 
 2 
 o 
 s 
 
 < 
 a 
 
TO MY TEACHER, 
 
 PROFESSOR CHARLES C. TORREY, 
 
 IN APPRECIATION OF HIS CONSTANT INTEREST 
 
 AND UNTIRING AID 
 
CONTENTS 
 
 PAGE 
 
 Preface 9 
 
 Introduction 11 
 
 The alternative readings indicated by the Qere and 
 
 Ketib 11 
 
 Alternative readings not noted in the margin by the 
 
 Massoretes 16 
 
 Some general rules that are found to have guided the 
 Massoretes in their method of combining alterna- 
 tive readings 21 
 
 Cases of Special importance 23 
 
 Alternative readings in the First Book of Samuel ... 26 
 
 Alternative readings in the Second Book of Samuel. . 45 
 
Pref 
 
 ace. 
 
 ERRATA. 
 
 Page 18, line 4 QC& should be Dn& 
 
 18, 
 20, 
 
 " 24 
 " 10 
 
 hi 
 
 mo 
 
 TT 
 
 20, 
 
 " 16 
 
 bx " 
 
 " ^ 
 
 23, 
 
 " 25 ni&b&n " 
 
 " Q^zhtfn 
 
 24, 
 
 " 17 
 
 ?]NS1D 
 
 " ?|N21D 
 
 24, 
 
 " 32 
 
 nw : 
 
 nssi 
 
 40, 
 
 " 3 
 
 T V 
 
 " in** 
 
 T V 
 
 52, 
 
 " 15 
 
 afcfl 
 
 2^ 
 
 54, 
 
 " 16 
 
 
 " my 
 
 — r 
 
 59, 
 
 " 23 
 
 nt£6#n " 
 
 t : — 
 
 mooouicviv uiai^lll lis SIXCIIl/. 
 
 That certain general rules have been followed in the 
 making of these combinations. 
 Baer's edition of the Massoretic text of the Books of Samuel 
 has been used as the basis of this work, but a few cases also of 
 
 1 1 am greatly indebted to Professor Torrey for his interest in this investi- 
 gation and for his kindness in giving suggestions that have proven invaluable. 
 
 Library Publications. 1. 
 
10 PREFACE. 
 
 Qere and Ketlb not included in this edition, but occurring in 
 certain manuscripts, have been discussed. 
 
 Wherever the terms "text" and "margin" have been used in 
 opposition they refer respectively to the consonant text and to 
 the Massoretic marginal note. 
 
 In addition to the passages discussed in the following pages, 
 the Books of Samuel contain a large number of cases of Qere 
 and Ketlb of a simpler and less interesting type, where the dif- 
 ference between the two readings is merely orthographic, or .due 
 to some accident of transcription ; or where a euphemistic substi- 
 tution has been made. In most instances it is evidently possible 
 to defend either reading. Thus for example "ITO^D and^n/Nt^ 
 I Sam. 1 : 17 ; 13*113 and ?31"13, 8:3; DH3 and DPD, 11 : 9- 
 
 : - t t : : 
 
 I have not thought it necessary to treat these cases here, though 
 they doubtless usually represent the combination of two manu- 
 script traditions. I give here a list of them, together with a few 
 passages in which it seems to me that the testimony of the 
 ancient Versions gives ground for supposing that our M. T. 
 includes more than one reading. The list is as follows : 
 
 I Sam. 1 : 17 ; 2 : 9, 10 ; 3 : 2, 18 ; 4 : 15 ; 5 : 6 ; 7 : 9 ; 8 : 3 ; 9 : 1, 
 26; 10: 7,21; 11: 6, 9; 13: 8, 19; 14: 27; 15: 6, 16; 17: 7,23,41 
 18 : 6, 7, 9, 14, 22, 28 ; 19 : 18 ; 21 : 12 ; 22 : 15, 17, 18 ; 23 : 5 ; 24 
 
 5, 9, 19; 25: 3, 18, 18; 26: 8, 22; 27: 4, 8; 28: 8, 24; 29: 5; 30 
 
 6, 24. II Sam. 1 : 8, 11, 16 ; 2 : 2 ; 3 : 2, 3, 12, 15 ; 5 : 1 ; 6 : 23 ; 8 : 3 
 11: 1, 24; 12: 9, 20, 24; 13: 8, 32, 33, 34, 37; 14: 7, 22, 30; 15 
 20, 28; 16: 8, 10, 12, 12, 15, 18, 23; 17: 12; 18: 8, 12, 13, 17, 20 
 19: 7, 14, 19, 41, 41; 20: 8, 9, 14, 23, 25; 21: 6, 9, 12, 16, 20, 21 
 22 : 8, 15, 23, 33, 34, 40, 51 ; 23 : 8, 9, 9, 9, 15, 16, 20, 20, 21, 33, 35, 
 37; 24: 10, 11, 14, 16,22. 
 
Introduction. 
 
 While the Qere and Ketlb in the Old Testament have hitherto 
 been considered to be mistakes or alleged mistakes corrected by 
 the Massoretes, a study of the marginal notes in the Hebrew of 
 the Books of Samuel — or, no doubt, of any representative por- 
 tion of the Old Testament — makes clear that this view must 
 be revised. The numerous instances also in these books of alter- 
 native readings which have not been commented upon by the 
 Massoretes throw valuable light on the question of Qere and 
 Ketlb and should be examined side by side with the Massoretic 
 margin. 
 
 I. THE ALTERNATIVE READINGS INDICATED BY 
 THE QERE AND KETIB. 
 
 1. At present commonly considered errors corrected by 
 the Massoretes. 
 
 It has been customary to translate the word Ketlb "that which 
 is written" and Qere "to be read" (Ges., §17, footnote: "Hp is 
 
 passive participle, legendum) . Kautsch in his Gramm. des Bibl.- 
 Aram., p. 81, note, translates Qere "das gelesene oder zu lesende". 
 Ginsburg, Intr., p. 183, says : "The Massorites have decided that 
 the marginal Keri is the correct reading", but does not otherwise 
 translate the word Qere directly. Caspar Levias, in the Jewish 
 Encyclopedia under "Masorah", paraphrases Qere "to be read". 
 Thus, the generally accepted view is, that the Massoretes, where 
 they discountenanced or considered incorrect the reading con- 
 tained in the text, placed a Qere in the margin against the word 
 or words in question and added there the reading that was by 
 them considered correct or preferable. The commentators have 
 treated the Qere and Ketlb in accordance with this view, and 
 speak of the Qere as a correction, now needed, now mistaken, 
 
12 INTRODUCTION. 
 
 and sometimes "unnecessary" (Driver, Notes on Samuel, the 
 note on II S. 12 : 22 ; H. P. Smith, II S. 12 : 24) . 
 
 The reason for this view of the Qere and Ketib is the fact that 
 the root fcOD means "to read" (out loud). It has been argued: 
 if the word that the Massoretes put in the margin was actually 
 read out loud at the reading of the Scriptures instead of the word 
 written in the text, it must have been considered preferable to 
 the reading of the text. But i"lp is the passive participle of 
 this verb, and must be translated "read", just as ^ip? is trans- 
 lated "written" ; but no tense or modal force must be ascribed 
 to it. 
 
 2. The use of the terms Qere and Ketib. 
 
 Qere may or may not have invariably been read in preference 
 to the consonant text; that cannot be determined; but several 
 circumstances can be brought forward to show that Qere and 
 Ketib were not always intended to distinguish between that 
 which was written in the text and that which was read out loud, 
 or should be read out loud. 
 
 Ginsburg, Intr., p. 187 ff., shows that in many cases the Qere 
 and the Sebir were interchangeable. Where one MS. had Qere 
 another would have Sebir, and vice versa. In fact, where one 
 MS. had Qere another would have Sebir, and still a third would 
 have a note to the effect that the Eastern reading was the same 
 as respectively the Qere and the Sebir of the first two. This 
 suggests that the three marginal notes, though called with dif- 
 ferent names, all had the same purpose — to record, beside the 
 reading in the text, an alternative. Furthermore it has been 
 shown that the Qere and the Ketib of one MS. have been reversed 
 in another, so that what is Qere in one is Ketib in another, and 
 vice versa; e. g. II S. 14 : 21, where in^J? and rVjyj? take turns 
 
 • • T T • T 
 
 as Qere and Ketib in different MSS. One is on this account led 
 to suspect that the words Qere and Ketib, if not originally arbi- 
 trarily chosen to designate alternative readings in a given pas- 
 sage, at least are very often used in such capacity with less of 
 their literal meaning. 
 
 a. Qere lost its sense of "read out loud". 
 
INTRODUCTION. 13 
 
 That the Qere had lost its meaning "read out loud", as distinct 
 from that which was "written", can be seen from the fact that 
 many words are supplied with a Qere in the margin, though they 
 would sound exactly the same whether read out loud according 
 to the textual reading or according to the marginal note. For 
 instance in I S. 2 : 3 t^lTOV ^50^ V?) "and by him actions are 
 
 weighed" is the Qere; and nl;?!? 1J3HJ tib) "though actions 
 
 be not weighed" is the Ketlb. JO, the negative, and 17, the pre- 
 position with the suffix 3rd pers. masc. sing., would, as far as 
 we know, sound alike in pronunciation. And while it may be 
 objected that the tone of voice of the reader would be different 
 in each case, depending on whether he understood the passage 
 according to the Q. or the K., it should be pointed out that in 
 both cases, if we are allowed to judge by our sense of emphasis, 
 the word in question would have the strongest accent of the 
 whole clause. Another instance is I S. 25: 8, where we have 
 !)J3 in the text. It is in some MSS. the subject of a marginal 
 
 T 
 
 note which tells us that the Qere is ^fcO- It is we H known and 
 
 T 
 
 generally admitted that ft quiesced in many cases, becoming 
 "merely a sign of the preceding long vowel" (Ges. 23 a ) . In this 
 case then, the Qere does not direct the reader to pronounce the 
 word in any way different from that of the Ketlb, for even with- 
 out the marginal note, and in all likelihood even without the vowel- 
 points, the readers for whom these MSS. were intended would 
 pronounce the vowel following the 2 Just as long as if they saw 
 the N actually written in the text. The cases of the quiescing N, 
 where the text lacks it and the margin supplies it, are quite 
 numerous; e. g. I S. 28: 24; II S. 2: 2; 19: 14; 20: 9; etc. They 
 furnish convincing proof that the distinction made by the Mas- 
 soretes between Qere and Ketlb was one that cannot be ade- 
 quately expressed by the translation of the words, "that which 
 is written" and "that which is read". 
 
 b. Ketlb is not always "that which is written" (namely, "writ- 
 ten" as the standard consonant text), but sometimes an alter- 
 native reading which makes sense, and which can be found by a 
 study of the details of the text and the marginal note. 
 
14 INTRODUCTION. 
 
 While we can almost always be certain of what constituted 
 the Qere, both as to consonants and vowels, the Ketlb is more or 
 less an unknown quantity. If we insist that the Ketlb must be 
 that which is written in the consonant text, exactly that and 
 neither more nor less, it will in many cases remain an unknown 
 quantity; for the consonant text very often brings together 
 letters into a combination which is impossible to vocalize or 
 classify as being one form or another of any Hebrew word ; e. g. 
 PIDDJ in I S. 15: 9; T\fcOm i n I S. 25: 34, etc. But if it be 
 recognized that by Ketlb was meant in most cases something that 
 makes sense and a possible construction, and that what is found 
 actually written in the text is a combination of the Qere and the 
 Ketlb, then the unknown quantity can almost invariably be found 
 by careful analysis of the seemingly impossible textual reading. 
 A good example of this kind of combination is found in II S. 
 
 19 : 32 HTQ HN ^7^7 " to conduct him over the Jordan" 
 (R. V.). The Qere, according to the margin, is j r n^n HN> the 
 
 same construction as we find in vv. 37 and 40. The Ketlb, if we 
 mean that which is written in the consonant text, is "destitute 
 of all philological analogy, and, in fact, meaningless" (Driver, 
 Notes on Samuel, ad loe.) . But Driver recognizes that we have 
 here "a mixture of two readings", giving one as JTV3 as in v. 
 
 19, the other one being the Qere. G B has iK-ire/xif/cu airbv rbv 'lopSdvrjv, 
 and G L has 7rpo7re'/i</"" airbv Zk tov 'lopSdvov, thus substantiating the 
 supposition that two traditions existed. That the present Hebrew 
 text in this passage is not an accident or a mistake which has 
 been corrected in the marginal note, but rather an intentional 
 preservation of two different readings in the MSS., seems beyond 
 the possibility of contradiction. 
 
 Did the Massoretes by calling n^PnTlN the "Qere" wish to 
 
 show their preference for this reading? That is a question which 
 is hard, if not impossible, to answer. But it is to be noted that 
 the reading of the two which most closely resembles that which 
 is written here happens to be given as the Qere. 
 
 Another passage worthy of note in this connection is II S. 5 : 2, 
 
 where the text has nrVPl iWN 13*6x7 t6d b)X& nVTia 
 
 - T • T T - " T Jv V T : • 
 
INTRODUCTION. 15 
 
 ^"T^TlN "OSrn N^B- If we consider the consonant text 
 here as it stands, it becomes plain that the construction is im- 
 possible, because "the words are wrongly divided" (Driver, ad 
 loc). The Qere therefore directs us to connect the final n of 
 nn^n with the following word, making ^alDH determined as 
 well as " l 2Qn, which would give us correct Hebrew. Butthe H could 
 very well be left at the end of the word nrPTl- We have several 
 instances of such full writing; e. g. I S. 15 : 3; II S. 2: 26 (Cf. 
 Ges. 44*0 . And the construction would be in harmony with 
 Hebrew style just as well, if both the participles, N^IlD and 
 
 "DDi were undetermined. We would then have a compound verb 
 form, rpH with a participle, denoting customary action in past 
 
 T T 
 
 time (Ges. 116 r ). Of course, in this case the J~| preceding "Oft 
 
 would have to be deleted. Without the Qere, however, either of 
 the two remedies would suggest itself just as readily. There 
 would be a shade of difference in meaning between the two 
 constructions. The Qere is : "thou wast the one leading out and 
 bringing in Israel". The alternate reading would mean: "thou 
 wast leading out and bringing in Israel", or with more emphasis 
 on the customary action : "thou wast wont to lead out and bring 
 in Israel". Here, too, if we take Ketib in its literal meaning, 
 "that which is written", it makes no sense ; and the construction 
 called the Qere is about as nearly like "that which is written" 
 as is the reading which constitutes the other element of the 
 combination. 
 
 II. ALTERNATIVE READINGS NOT NOTED IN ^THE 
 MARGIN BY THE MASSORETES. 
 
 Examples of the kind mentioned above, where the consonant 
 text itself, even without the vowel points and without the Mas- 
 soretic margin, bears plain evidence of combination, are very 
 valuable for the argument of this investigation, inasmuch as 
 they furnish a link between the class of alternative readings 
 commonly called Qere and Ketib and the great mass of cases 
 
16 INTRODUCTION. 
 
 where two different readings have been combined without any 
 marginal explanatory note. The two passages just cited differ 
 from the majority of the cases of Qere and Ketib in this respect, 
 that here the two combined readings are plain to the reader 
 without the marginal note, while in most cases where the Mas- 
 soretes have given aid the note or the vowel points or both are 
 necessary to show that combination has taken place: the con- 
 sonants alone cannot make this apparent. A great many cases 
 of this most common kind of Qere and Ketib will be discussed 
 in the following pages and no special examples need be pointed 
 out — they are found in almost every chapter of the two books 
 under discussion. But in other places the text bears unmis- 
 takable evidence of combination and furnishes certain clues to 
 the component readings, while the margin is silent. This class 
 of alternative readings has not hitherto been generally recog- 
 nized, though it is fully as important as the Qere and Ketib and 
 is similar in purpose to these. 
 
 1. Cases where the alternative readings have been put 
 into a combination sufficiently intelligible to make a 
 marginal note unnecessary. 
 
 By comparisons of examples representing the two classes of 
 alternative readings, those noted in the margin by the Masso- 
 retes, and those not so noted, the following becomes plain. When 
 a MS. was copied, the copyist often had recourse to two or more 
 different readings of certain passages in the book he was to 
 copy. Whether or not he always decided in his own mind which 
 of the different readings he preferred, is of course impossible 
 to say. But what he often did was this: he combined two, or 
 perhaps sometimes more, readings in the text. Where this com- 
 bination could be affected in such a way as would make clear 
 to the intelligence of the average reader the two different read- 
 ings combined, the copyist would not as a rule make a marginal 
 note of Qere. On the other hand, where such a combination 
 could not be made, and where consequently one of the readings 
 would otherwise be lost to the reader, the copyist would as a 
 rule make a marginal note, which together with the text itself 
 should preserve the two readings. Of course, it should be re- 
 membered, that some of the cases of Qere and Ketib may be due 
 
INTRODUCTION. 17 
 
 to errors found in the text and corrected by the Massoretes, but 
 these cases are greatly in the minority, and can be called excep- 
 tions to the rule. 
 
 In support of the preceding paragraph the following is found : 
 
 i s. is: is nnpnnni ~r\b nptfn Tpi? nirr r^nhm 
 onfc antes iv In nbrfoi pb6v n« crabnn h«. 'ft is 
 
 t t - - t : - : ■ : I ■■ r —. :• • t 
 
 evident that the text as it stands cannot be correct. But the 
 clause as a whole is so ordinary and the meaning of it so plain 
 from the context, that the solution of the problem is quite 
 apparent. After having read: "And Jehovah sent thee on a 
 journey, and said, Go, and utterly destroy the sinners the Amale- 
 kites, and fight against them until . . . ", then seeing the words 
 Dn& Dnte3 "11?> one evidently containing the Piel of the root 
 
 T T - - 
 
 7w2 "to finish" or "to consume", the other including the suffix 
 3rd pers. plur., one will most naturally end the sentence either 
 "until they be consumed" or "until thou consume (or finish) 
 them". But it is not only the context that in a general way 
 suggests what each of the two readings is. By carefully noting 
 every detail of the combination one will arrive at the same solu- 
 tion of the problem. Having read the two words Dni?S "117, 
 
 T - - 
 
 every one acquainted with the Hebrew idiom would feel that 
 the sentence is finished, the sense is complete; and, since the 
 superfluous 2 p^ adds no new thought, he would put a period after 
 
 DnlyD TU- Perhaps he would recall passages such as I Kings 
 
 22: 11; Ps. 18: 38. And yet there is DnN. written in the text 
 
 T 
 
 but not accounted for in his first translation. Calling to mind 
 such Hebrew constructions as Dn& TlteS "II? in Jer. 9: 15 and 
 49: 37, he would know that the reading Qnfc "Hntes IV 
 
 would be suitable here, just as well as anyone somewhat versed 
 in English would know that either "until they be consumed" or 
 "until thou consume them" would be a suitable conclusion to 
 the English sentence. The two alternate readings thus arrived 
 at we find represented in the Versions. LXX, Syriac, and Targ. 
 have Dn& "nntes IV, while Vulg. has Dntes IV- There is 
 one more detail which should not be left unnoticed. The reading 
 
18 INTRODUCTION. 
 
 □JIN n1?3 IV would be possible in Hebrew, and would make 
 sense here. But the suffix 0_ of Qn1?5 serves the double 
 purpose of designating the object of the inf. HP 3 in the one 
 reading, and in the other, where we retain Qc& as the object, 
 to suggest that another suffix was added to the inf. pfp3 as 
 
 subject (cf. rule 4, p. 21). That no other suffix would do as 
 subject except that of 2nd pers. sing, is plain from the context. 
 If it had been intended to combine the two readings Dn1?3 IV 
 and Qn& n1>>3 1V> ^ i s natural to suppose that the combina- 
 tion would have been □!"!& ni>0 IV (similar cases will be 
 
 seen in the sequel), and probably with a Qere, since the con- 
 sonants alone, if this were the text, would not clearly indicate 
 what two readings were combined. 
 
 is. 15:9 iDnnn nm ddji hddj rD*6sn b^ "but 
 
 • v : v t ' • T : t : • : t t : - t : 
 
 everything that was vile and refuse, that they destroyed utter- 
 ly", R. V. One of the difficulties in the text here is the word 
 HODX which has been pronounced "a grammatical monstrum, 
 
 t : • : 
 
 originating evidently in the blunder of a scribe" (Driver, ad loc.) . 
 Explanations have also been offered to show how the scribe by 
 mistake began the word DDJ first, then noticing his mistake 
 
 simply added r\V2 with iTQDJ a s the result. 2 The probability 
 
 is, however, that the scribe intentionally wrote this word exactly 
 as it stands, intending thereby to combine the Niph. participle 
 nOJ. from nD "to despise", and the adjective n*D. from HID 
 
 "to suck out", "to exhaust" (cf . Deut. 32 : 24 "2V") "'IID "wasted 
 
 t t ■■ : 
 
 with hunger") . There was no other way of combining the two 
 except to put the two radicals Q and 5 side by side in the word ; 
 and as nothing else could be made of it, it was evidently expected 
 that the word would be correctly understood, even without a 
 Qere. The word has generally been read nOJ> which therefore 
 
 t : • 
 2 It is now generally recognized that in place of ( T|Hk* DP 31 we snouli ^ 
 read nDNDJl- 
 
INTRODUCTION. 19 
 
 needs no further comment. The sense of HTD, the alternative 
 
 TT 
 
 reading, is most excellently suited to the context. "The best of 
 the sheep, and of the oxen, and of the fatlings, and the lambs, 
 and all that was good" has just been spoken of, and the opposite 
 is now mentioned. What more appropriate word could be used 
 in this description than niD "emaciated"? 
 
 TT 
 
 These two examples, then, to which a great many others could 
 be added from those discussed in the following, show that the 
 very same thing has taken place here as in nearly all the pas- 
 sages marked with a Qere, namely: two readings have been 
 combined. The only difference is this, that in these cases no 
 marginal note was considered necessary, because the text itself 
 made the alternative readings sufficiently clear. 
 
 2. Cases where two different traditions are indicated 
 by the use of different words or constructions in close 
 proximity, while ordinarily identical words or construc- 
 tions would be expected. 
 
 There is every reason to believe that a scribe would try to 
 avoid marginal notes as far as possible (cf. Ochlah W'ochlah, 
 p. VII, 3) . If he could make some kind of intelligible combina- 
 tion without resorting to the margin, he would. For instance, 
 it is well known that the prepositions 7N and jy are very 
 
 often interchanged, as it would seem, without any difference in 
 meaning. In the Books of Samuel we have numerous instances 
 of Vn being used where we should naturally expect ~>y, and 
 vice versa. The following are worthy of note : I S. 14 : 10 l?y 
 Why, and 14: 12 U"6n ^7V- In the first of these two in- 
 stances some 20 MSS. read U^N )bv Just as in 14: 12. Our 
 text records two readings by retaining ?J? in one place and ?N 
 in the other in the very same context. I S. 25: 17 HTwD "O 
 
 t : t • 
 
 1rP3 b3 hi!) W^'-IN b# iljnn. Here, too, b# and by are 
 
 t ~~ : ■•-:••• t tt 
 
 used in the same sense and almost in the same breath. I S. 27 : 
 10 contains a similar case. But the most striking case of the 
 interchange of ^ and by is found in II S. 2: 9, where we 
 
20 INTRODUCTION. 
 
 have first 7^ three times and then ?}7 three times in the same 
 
 sentence. The case is equalled only in I S. 9: 4, where six 
 verb forms having the same subject are divided equally be- 
 tween the sing, and the plur. (cf. note on this passage) . There 
 was ample opportunity in sentences like these cited, where the 
 same preposition is expected two or more times, for the scribe 
 to indicate that one authority had 7^ uniformly, and another 
 7$, without resorting to a marginal note. Only once in the 
 two books under consideration are these prepositions combined 
 into one, with the consonants of 7^ and the vowel of 7$$ 
 
 Naturally, we find here a Qere, for the consonant text alone 
 cannot record both readings. But it is important to note in 
 this passage, that there was no opportunity to put both prepo- 
 sitions into the text except in this way. There is only one word 
 in the whole verse which could possibly be governed by either 
 7^ or 7^ in this connection. True, the next verse contains 
 
 alternately the two prepositions, but the idiom there is different, 
 and would for that reason not suggest the two readings for this 
 verse. Here it is 7>y 35^"' or QPi 7t"l 7#, in the following verse 
 bv "2VT 1 or DStfian b#- How else then could the double tra- 
 
 - T T 
 
 dition have been preserved in this verse, except by a marginal 
 note ? In this connection compare also the combinations of num- 
 ber in I S. 30 : 22 and in II S. 21 : 4, which passages are treated 
 in the following. The first of these two places has no marginal 
 note. The combination must have been thought to be sufficiently 
 obvious without it. 
 
 3. Cases where two different readings, commonly known 
 as doublets, have been incorporated in full into the text. 
 
 In certain instances the scribe seems either to have found it 
 impossible or deemed it impracticable to combine two readings 
 available, whether by one method or another of those described 
 above. The solution he then resorted to was simple: he incor- 
 porated both readings in full into the text — generally side by 
 side, though not always. IS. 19 : 20 ; 24 : 7 ; II S. 3 : 13 ; 5 : 1 ; 
 
INTRODUCTION. 21 
 
 22: 43 are examples of two readings put into the text side by 
 side, while I S. 17: 13 affords an instance of the two readings 
 included in full, but not side by side. 
 
 4. Cases where one reading only is in the text, but the 
 context is made to suggest another in addition. 
 
 There remains one more class of alternative readings to be 
 mentioned; those where, strictly speaking, only one reading is 
 found in the text, but where the context inevitably suggests the 
 alternative. In such instances, it seems, the less obvious reading 
 was put into the text, while the easier construction, the one that 
 would more readily be suggested by the context, was left to be 
 implied in the mind of the reader. Instances of this kind we 
 have in many places where either nPlN "thou" or HHU "now" 
 
 stands in the text and suggests each the other ; e. g. I S. 28 : 2 ; 
 II S. 7: 20.; 18: 3. The two cases in I S. 31: 9 of alternative 
 readings of this kind should be especially noted. 
 
 III. SOME GENERAL RULES THAT ARE FOUND TO 
 
 HAVE GUIDED THE MASSORETES IN THEIR METHOD 
 
 OF COMBINING ALTERNATIVE READINGS. 
 
 It is a well known fact that the Massoretes had a critical 
 apparatus of their Holy Scriptures, and this investigation may 
 in some measure prove that they also used this apparatus in 
 their transcription of the MSS. Their desire to include valu- 
 able variant readings supplied by the critical apparatus prompted 
 them to make combinations such as have been described above. 
 The following are some general rules, that seem to have guided 
 them in doing this. 
 
 The one rule superseding all others seems ever to have been 
 based on the necessity of making the combinations clear. Fur- 
 ther, the peculiar circumstances of each combination made a 
 rule which, of course, in that particular case had to be followed. 
 In other words, whatever was most convenient was the proper 
 method. These two were ironclad rules. Those that are men- 
 tioned below are more flexible, and would therefore have to 
 
22 INTRODUCTION. 
 
 yield to the two mentioned above in every instance where they 
 would conflict. While this must be borne in mind distinctly, the 
 following paragraphs may in some measure be found instructive. 
 
 1. Wherever possible, marginal notes should be avoided. The 
 Scriptures were considered holy by those who had in charge the 
 work of copying them. The scribes realized that when a mar- 
 ginal note was once put into the MS. it would stand there as a 
 human element in the midst of the divine, and would very easily 
 lead to grave complications of the text. See also pp. 16, 19. 
 
 2. The consonants of one reading and the vowels of the other 
 should be put in the text, wherever such a method would make 
 the combination clear. This principle has been known and ex- 
 plained before by different scholars (cf. Ginsburg, Intr. p. 183 f.) 
 and little need be said here. Illustrations of it are numerous 
 in Samuel as well as in any other part of the Old Testament, 
 but the following examples, discussed in the following pages, 
 will suffice : I S. 10 : 7 ; 13 : 8 ; 15 : 16 ; 20 : 24 ; II S. 3 : 2 ; 13 : 8 ; 
 14: 22; 23: 8. 
 
 3. A consonant not common to both readings should not have 
 any vowel, if put into the text ; and if it is not put into the text, 
 a space should be left for it there, and its vowel placed under 
 the space. Illustrative examples of this rule are found in I S. 
 14: 32; 20: 2; 24: 5; 26: 8; 26: 22; II S. 13: 33; 15: 21; 19: 
 41; 21: 9. 
 
 4. The number of consonants in a combination should, where 
 it is possible, be made to correspond with the reading which is 
 represented by the consonants, even though the reading indi- 
 cated by the vowels should require more or less. While, just 
 as the other rules, this one can not be said always to be followed 
 strictly but is made to yield to whatever is found most con- 
 venient and most intelligible in each particular instance, yet 
 there seem to be some cases where it plays an important part 
 in solving the combination. Such are I S. 2 : 10 ; 15 : 18 ; II S. 
 23 : 8 a . The notes on these passages should be compared in this 
 connection. Of especial interest is the marginal note against 
 "•nfcOni in I S.. 25 : 34. The word in the text has been pro- 
 nounced by Nestle a combination of i&Om and flfcOT (ZAW, 
 
INTRODUCTION. 23 
 
 XIV, p. 319), and this has been accepted by many scholars (cf. 
 Ges. 76 h ). But the margin reads n&iDnii which seems to be 
 
 : t~ 
 
 capable of no satisfying solution, unless it is understood as an 
 
 attempt to improve on the combination found in the text. What, 
 
 then, could have been the objection to TlfrGPll ? It was evidently 
 
 : t - 
 
 this, that TlfrOm would stand for ">X3m o,nd TUO 1 !, the 1st 
 
 : t- t- • T 
 
 pers. sing, of the perf. As it is, the context makes the misunder- 
 standing impossible but the author of the marginal note seems 
 in this case actually to have corrected the combination already 
 found in the text. In both the readings, ""fcOm and HfcOl 
 
 t - : t 
 
 only one consonant as ending follows the last radical of the verb, 
 K; and by having iJliOm in the text the presence of both p 
 
 T - 
 
 and ^ at the end is misleading. In Qere, vocalized as the impf., 
 the H holds the place of "> of i&QM, while it at the same time 
 
 is the ending of the 2nd fern. perf. Of course, the J"| at the 
 beginning had to be there as the representative of the impf. 
 The perf. is more easily suggested by the present form TliOFn, 
 
 : t _ 
 
 in spite of the preformative, than the impf. could have been, had 
 the preformative been lacking. 
 
 The situation in II S. 23 : 8 is also illuminating. The two 
 readings combined there are WpbWn £>N"I and TWlhzTs VMl. 
 The text is vocalized as though the word were D^EvETU but 
 
 • • T - 
 
 the final Q is excluded, in order that the number of consonants 
 may indicate the alternative reading as riDvG^n and not 
 U^bWn, as is the case in I Chr. 11: 11; 12: 19 (where the Q 
 is put in) . 
 
 IV CASES OF SPECIAL IMPORTANCE. 
 
 The marginal notes in I S. 25: 34 and in II S. 3: 25 are of 
 incalculable value as evidence more directly from the Masso- 
 retes themselves of the fact that they consciously and inten- 
 tionally put seemingly impossible forms into the MSS., in order 
 thereby to preserve the different traditions found in certain 
 passages. I S. 25: 34 has already been discussed above, and 
 
24 INTRODUCTION. 
 
 what has been said there will probably suffice, except that stress 
 should be laid upon this : if we consider the form written in 
 the text a "monstrum" (as it was once considered), that would 
 reflect unfavorably upon the conscientiousness or intelligence of 
 the Massoretes, who had in charge the copying and editing of 
 the MSS. ; the mistake could, however, be ascribed to universal 
 human fallibility; but when we also find in what we term their 
 "correction" of the erroneous word in the text a form on the 
 face of it just as impossible grammatically as the form it was 
 intended to correct, then we are obliged not to utterly discredit 
 the grammatical intelligence of the scribes, but rather to seek 
 a satisfying explanation for the peculiar phenomenon. The ex- 
 planation given in the note on I S. 25 : 34 and in the preceding 
 paragraph seems to satisfy in every detail, and the present 
 writer for that reason feels reasonably assured that it is correct. 
 The case in II S. 3 : 25 is very similar. The text here contains 
 the quite ordinary expression TjitfDD riNI ^N^ID JIN "thy 
 going out and thy coming in". The consonant text is exactly 
 as we should expect it, and if it were not for the vocalization 
 and the margin, there would be no problem to explain. But the 
 Qere is TjNDlO. found only in Ez. 43: 11 besides here, and 
 
 termed "abnormal and incorrect" by Driver (Notes on Sam., 
 ad loc.) . In the Ezekiel passage the margin is silent, in spite 
 of the fact that the text has this "abnormal" and "incorrect" 
 form. If we are prone to discredit the scribes, here is an ex- 
 cellent opportunity to accuse them of doing the wrong thing at 
 the wrong time. Where the text is right, they seem to have 
 made it incorrect, and where it needs correction, they have left 
 it as it is. Both Driver and Konig (II, p. 127) think that per- 
 haps for the sake of assonance the Qere has been vocalized as 
 it is. But there is a more plausible explanation. The Qere here 
 and the similar form in Ez. 43: 11 TlfrOift i"lX1 TlNUlD HN 
 are intended to show that some authorities had TjNlJlD first, 
 and others TjNlDD- The fact that TjNlDTp is vocalized as TjNalD 
 was meant to indicate this. But this theory is supported not 
 only by the fact that it sounds plausible and is brought forward 
 with a mass of similar examples ; we have also testimony from 
 the Versions that authorities did exist which had the words 
 
INTRODUCTION. 25 
 
 NylD and NiDD in reversed order. In the passage here under 
 
 T "" T 
 
 discussion the Lagarde text of LXX has: ttjv dVoSo'v a-ov ko.1 t^v 
 c^oSov o-ov, representing the alternative reading, while the Syriac 
 of the Ezekiel passage has the word-order of the Hebrew re- 
 versed. 
 
 Library Publications. 2. 
 
Alternative Readings m the First Book 
 
 of Samuel. 
 
 CHAPTER II. 
 
 3. nlV^y 1J3nJ tibl Ketib "though actions be not weighed". 
 Qere fll^i? U3nJ V?) "and by him actions are weighed". 
 
 Either construction is possible, and represents a separate reading. 
 Greek favors Q. Syriac has a mixture of both readings perhaps. 
 The context favors Q. The adoption of Ketib would mean a 
 weakening anti-climax, which, however, could be defended. At 
 any rate, Q. is the easier construction and gives the expected 
 
 meaning. This is quite significant; for though fcO in the text 
 
 would suggest to the reader *p, the reverse is not true. It is 
 
 possible that this is the reason why JO and not *Q was put in 
 the text. 
 
 10. nnD iniT Him- Ketib tnD, Qere mnD "They 
 
 T • : — t : • : . t • : 
 
 that strive with Jehovah shall be broken to pieces" or "he that 
 striveth". Both readings are equally well suited to the meaning 
 of the context, though, in case K. be adopted, the previous word 
 would have to be made to correspond with the sing. ^"ID- 
 Greek favors the sing., and Syriac the plural. But the choice 
 here of Q. or K. depends largely on the interpretation of the fol- 
 lowing word, which see. 
 
 10. □J?"V [TD$3 )hV- K etib rib]}- Q erS VbV- Ketib trans- 
 
 ••:-•- T - T T TT TT 
 
 lated: "He ascended in the heavens thundering" (cf. Greek). 
 Qere: "Against him will he thunder in heaven", y}]) is written 
 
 T T 
 
 thus only here and in II S. 20 : 8, which see. This unusual form 
 therefore cautions the reader that another word of similar sound 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 27 
 
 and with the same number of letters should be substituted in 
 the Ketib. The only word that fulfils this condition and fits the 
 context is rVV- It is quite interesting to note that after ex- 
 
 T T 
 
 eluding from the Greek of this verse the part that has crept in 
 from Jer. 9 : 23, 24, we have left quite a faithful representation 
 of the Ketib throughout the verse, even including the sing. I^^D 
 
 (see preceding note) . It is also to be noted that Syriac repre- 
 sents Q. throughout the verse, having even 3rd pers. plur. suff. 
 after the preposition b>V- Our text could not very well have 
 preserved that feature too, since Drp7l? would not have lent 
 itself to combination with rOV- But of course in the mind of 
 the reader who adopts V^'ID) )/V stands for DrivJJ- There 
 
 t • : t t — : 
 
 are then, strictly speaking, three different readings combined in 
 this verse, and all are possible and suitable in the context; the 
 choice is with the reader. It may be doubted whether the emen- 
 dations now generally adopted here (see the comms.) are de- 
 sirable. 
 
 16. jnn nny "0 )b "IDN1. Ketib "and he would say to him: 
 
 Thou shalt give". Qere fnn r\P\V "G tib "1DN1 "and he would 
 
 I •• • t - • - t : 
 
 say, Nay, but thou shalt give". Either reading is perfectly clear, 
 though perhaps Q. is a little more vivid. 
 
 CHAPTER IV. 
 
 13. PIBSD Tj"PI "Jl NEOn by 2^ "^5? "Eli was sitting upon 
 his seat by the wayside watching", R. V. Ketib 7p; Qere "p. 
 
 The Ketib makes no sense, and has generally been considered a 
 clerical error. It is very probable, however, that the lengthening 
 of 1 to "I in this word was intended to serve a definite purpose, 
 
 namely to warn the reader to stricter attention to the details of 
 the text, so that he might see that the consonants PlOUQ mav be 
 read either ("IDS?? "watching" or HSifP " M ispah". Thus two 
 readings are possible here by changing the pointing: one, "Eli 
 was sitting upon his seat by the wayside (read THin) watch- 
 
28 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 ing" ; the other, "Eli was sitting upon his seat by the side of the 
 road to Mizpah". 3 The "j of the Ketlb is a part of neither of 
 
 these two readings, but is put in the text as a "Nota bene !" or 
 "Obs. !" Several other passages can be cited where something 
 out of the ordinary in the text has a function similar to that 
 of the K. in this verse. See for instance Jud. 16 : 5, where the 
 pathah in jpg is plainly out of place. IfO or fanJ is what we 
 should expect. But if either of these two forms had been written 
 in the text, the other would not have suggested itself. As it now 
 is, both of them will naturally come into the mind of the reader. 
 Note also the small n in Gen. 2 : 4, the suspended J in Jud. 18 : 
 30, and the large and the small D in Dan. 6 : 20, in all of which 
 places the extraordinary appearance of the text is intended to 
 direct the reader's attention to two possible readings. 
 
 is. ny^'n t -13?? n^nhx ndsh bya hm_ "he feii from 
 
 off his seat backward by the side of the gate", R. V. In this 
 translation "ll?3 is not represented. Driver (ad loc.) suggests, 
 
 that we should probably read "llJ^n "1172 "through (an opening 
 
 in) the gate". The text as it stands cannot be translated and 
 give satisfactory sense. But the two translations, the one sug- 
 gested by Driver and that of the Revised Version, are both pos- 
 sible and satisfactory, if we in the case of the one delete ~p, 
 
 and in the other "1173. The fact that both these translations 
 
 have offered themselves to different readers shows how effective 
 was the method used in combining the two readings "ly&'H "1173 
 
 and ~ll?£''n "P- In the latter reading ~p is construed as ace. 
 
 of place. 
 
 CHAPTER VI. 
 
 4. Dr^D^ D^b nnN HD^D "O "for one plague was on 
 
 ....... T : : t- - 
 
 them all, and on your lords", R. V. margin. Evidently, here is 
 a mixture of 2nd and 3rd person. Greek and Syriac have 2nd 
 person uniformly. Thus was perhaps the most common reading. 
 
 3 The Heb. text rendered by LXX had suffered contamination from vs. 18. 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 29 
 
 But with this reading has been combined one which had 3rd 
 person throughout. Both these constructions are possible. In 
 one case, the clause is included in the preceding quotation, in 
 the other, it is to be taken as an explanation by the narrator. 
 Note, that a marginal Qere was unnecessary here, since there 
 were two words that required suffixes. The one could be supplied 
 with a 2nd pers. suffix, and the other with a 3rd pers. suffix. 
 
 CHAPTER VIII. 
 
 19. irbl? n\T rf?D DN i? tfh nptf'l "And they said, Nay, 
 but we will have a king over us", R. V. This passage recalls 
 the one in 2: 16, where, after the verb "")PN, "3 j? was found 
 in the text, while the margin substituted ^7. The possibility 
 suggests itself at once of reading *p instead of JO, and the 
 dagesh in the $) is the key to the situation. For, reading the 
 consonants as they stand, one would naturally make a short stop 
 
 after ! Hp&0% and let fcO begin the quotation; and that is one 
 acceptable reading. But the dagesh indicates that #7 should be 
 closely connected with lipfcOl and be pronounced something like 
 1y*ipfc0l- Of course in this case the ^> must be changed into 
 
 )?,' since this alone of the two would be capable of such close 
 connection with Tip^l- We have thus, by the pointing of the 
 
 Massoretes, arrived at an alternative reading, Qfc$ *3 )y 'Hpfc^l 
 
 "and they said to him, But, etc.", which is equally acceptable, 
 as regards both construction and meaning. The Greek too com- 
 bines, but leSS Cleverly: Kal el-rrav avrcS Ovxl, a\X 7) /3a<TiXeu's, etc. 
 
 Syriac has: Ifl H** P °r^l°- Cf. 10: 19 and 12: 12. 
 
 For similar cases of dagesh in #j, cf. Gen. 19 : 2 and Hab. 2 : 6. 
 
 CHAPTER IX. 
 
 1. "^P 1 D' n N D " the son of a Benjamite". ijipi $t# is 
 found elsewhere, II S. 20: 1; Est. 2: 5, but the combination 
 "O^D" 1 E^K |3 occurs nowhere else, and is here highly improb 
 
30 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 able. If the father of Aphiah was not known by name (Driver, 
 ad loc), it is not likely that an attempt would be made to con- 
 tinue the genealogy by stating that Aphiah was "the son of a 
 Benjamite". The genealogy would be more likely to end with 
 a definite statement as to the tribal connection of the man last 
 named. Either ijipi n or "^p" 1 K^N is what we would expect 
 here. And the fact that Lucian's Recension of the Greek and 
 the Syriac express only ijipi 0^ makes it reasonably certain 
 
 that we have here two readings, ijip") |3 and ^""p 1 &X, both 
 included in the Hebrew text. 
 
 3. D^SMinQ "1!"!^ "one of the servants". While the construct 
 state before a preposition in Hebrew seems to be of quite fre- 
 quent occurrence (Ges. 130 a ), it may be well to note it here 
 because of a similar case in II S. 10 : 9, where the Massoretes 
 have made a marginal note directing the reader to delete the 
 preposition. It is perhaps hard to determine whether this case 
 is due to the effort of the scribe who desired to combine the two 
 readings, D^WH "DIN and D"Hl?Hnn "UlN. or whether our 
 present text is original. 
 
 4. ^^D""lby , l"nDU , V'^^D""lbn"lbPT six verbs, all 
 
 t t -: — : t t -: — -: — 
 
 with one and the same subject, though three are in the sing, and 
 three in the plur. They evidently represent two different tradi- 
 tions; one with the sing, throughout, . the verbs then referring 
 to Saul alone as the subject, without taking account of the 
 servant with him; the other with the plur. throughout, with 
 Saul and his servant as the plur. subject. Note that the verbs 
 are divided equally between the sing, and the plur., seemingly 
 without preference expressed for one reading or the other. (Cf. 
 II S. 2: 9, where 7^ and 727 are distributed in the same way.) 
 
 The Greek has here the plur. uniformly, while the Syriac renders 
 first two verbs in the sing., then one in the plur., then two in the 
 sing., then one in the plur. The combination here becomes suf- 
 ficiently evident to the reader without any marginal note. 
 
 24. rrbym pwn na nacsn dti "And the cook took up 
 
 the thigh, and that which was upon it", R. V. Though such a 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 31 
 
 translation makes possible sense, it does not suit this context. 
 Driver (ad loc.) doubts strongly that this punctuation represents 
 a genuine tradition, and adds: "The usage here (of H as a 
 relative with the prep, hv) is exceptional, and entirely unsup- 
 ported by precedent or parallel." Under the circumstances it 
 can scarcely be doubted that Geiger (Urschrift, p. 380) is right 
 in reading n* l 7^m "and the fat tail". The context is particu- 
 
 t: - T : 
 
 larly well suited by this reading, the allusion in the verse being 
 "to certain choice pieces reserved specially for those honored 
 with a place D"Wlpn DN"D" (Driver, ad loc). Only one 
 
 change in the consonants of the text is required, namely ^ 
 substituted for y. But how account for the y? The vowels, 
 moreover, as well as the substituted consonant, suggest a verb 
 with the 3rd fem. sing, suffix, and the context determines the 
 form: rPSTl- This would mean "and he brought it up", assum- 
 ing that the cook advanced to the place where the most honored 
 guests were sitting, from the "lower" end of the "guest-chamber" 
 where the food was being prepared. This second (perfectly 
 plausible) reading was combined with the other. Q*n can i* 1 
 
 VT" 
 
 either case very well be translated as above, "and the cook took 
 up", i. e. the thigh from where it had been lying. 
 
 CHAPTER X. 
 
 19. Why D^n T\hD ">3 )h nptfni "And ye have said 
 unto him, (But) set a king over us". "Q may be regarded here as 
 
 only the sign of a direct quotation following. But the fact that 
 " l 3 so often is used after the negative, in the sense "but", sug- 
 gests to the reader the possibility of reading fcO instead of )y, 
 thus translating the sentence : "And ye have said, Nay, but set 
 a king over us". So the Greek, Syriac and Latin versions ! Cf . 
 8: 19 and 12: 12; also on 28: 2. 
 
32 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 CHAPTER XII. 
 
 12. why "fin] "]bp "G i/6 ^b np^ni "And ye said unto 
 me, Nay, but a king shall reign over us". The interesting feature 
 of this passage is that it combines, or rather includes in full, the 
 two different readings found in 8: 19 and 10: 19. Greek B and 
 Syriac exclude "17. Greek L agrees with the Hebrew. 
 
 13. DV)bxp 1E'N Drnn? *I#N rfrtpn "The king whom 
 ye have chosen, whom ye have asked for". This is a doublet : 
 two different readings, both included in full in the Hebrew. 
 Greek B lacks □rP^tp' *"l£^- Greek L and Syriac support the 
 Hebrew. Either of the clauses would render the sense complete. 
 "/NC" is used of the request for a king in 8: 10. Nevertheless 
 Drp^^' "l^N appears here to be superfluous, and is probably 
 to be omitted with LXX." (Driver, ad loc.) . 
 
 CHAPTER XIV 
 
 7. 7]b HCpJ ^^7^ ~\p$ ^ HEW "Do all that is in thy 
 heart: turn thee", R. V. The sense thus rendered is not very 
 suitable, especially in view of the fact that the usual force of 
 mDJ is "incline", which is undesirable if the rest of the text is 
 
 . TT 
 
 kept as it stands (see Driver, ad loc). Greek B has: nout wav S 
 iav 17 KapSla aov IkkXCvyj, and L has the same, except that it sub- 
 stitutes OeXrjar] for IkkXivtj, and adds at the end of the verse: 
 ZkkXwov o-eavTov, These two authorities, therefore, point to a 
 
 reading: ^ niDJ Tp^? 1p# blD HUV "do all unto which thy 
 heart inclines". This construction is plausible here, and the 
 idiom has parallels: Jud. 9: 3; Ps. 119: 36, 112. But the prep. 
 3 before TpD 1 ? suggests another expression, equally idiomatic 
 and suitable: TjD^3 1&$ ^3 nfcV "do all that is in thy 
 heart". Cf . Ex. 35 : 34 ; Ezr.' 7 : 27. The Syriac has this reading 
 represented twice in the verse. It seems therefore reasonably 
 certain that our text combines the two readings. Why )b should 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 33 
 
 have been changed into TH is perhaps hard to say, but it may 
 
 be supposed that the change was due to some scribe's failure to 
 understand that the text contained two readings. 
 
 10 and 12. ^y ^y and IJ^N )bv "Come up to us". 
 The use of ?y and 7^ in the very same expression and in 
 such close proximity is probably an attempt to show that some 
 MSS. had ?y in both cases and some ~>$. See introduction, p. 19, 
 andcf. 25: 17; 27: 10; II S. 2: 9. 
 
 32. 77$° 7N DVn fr]?"! Ketib 77^' Qyn &V*1 "And the 
 
 XT" V TT — " TT TT 
 
 people made spoil". Qere 77^'n 7^ D17H CDITI " An d the peo- 
 
 T T - V T T 
 
 pie flew upon the spoil". Q. is plausible and very descriptive, 
 and deserves preference rather than K. as the original reading. 
 However, K. also is idiomatic and has support in passages like 
 Gen. 12: 15; 31: 1; Jer. 17: 11; I Kings 1:5; Deut. 8: 17, 18. 
 It is to be noted that a part of the one construction and a part 
 of the other were admitted into the text. The incongruity of the 
 preposition 7^ following nt^S? w &s considered sufficient indica- 
 
 T T 
 
 tion to the reader that it must be dropped in case K. is adopted ; 
 while the presence of the preposition 7^ and the similar phrase 
 775^'n 7$, £Dl?ril in 15 : 19 could be counted on to suggest the Q. 
 It is probable, however, that jyiPI originally came into the text 
 
 as an error of transcription and that the sequel was changed to 
 make the context grammatically consistent. This is a very in- 
 teresting case. 
 
 CHAPTER XV. 
 
 1. bynty) by lSy bv Tlb^h "To (be) king over his people, 
 over Israel". B has only m 'Io-pa^A., and L only iwl rbv Xaov airov 
 Syriac has both phrases but in reversed order. We have here 
 two different readings, both included in full in the Hebrew text. 
 
 1. nliT "Hin 7lpb VD& "Hearken unto the voice of the 
 words of Jehovah". B and Vulgate have only Hi IT 7lp7 VD& 
 
34 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 represented. L and Syriac correspond to the Hebrew. It may 
 be that here too the text combines two different readings in full 
 
 n|rr b)pb yp& and nliT nrn ynp. 
 
 9. DCil niDDJ " Vil e and refuse", R. V. ,~IDDJ combines 
 
 ••T : t : • : t : ■ : 
 
 the Niph. participle of HD and an adjectival form of HTD; 
 
 TT TT 
 
 i*inj "despised" or "vile" and HID "sucked out" or "wasted 
 
 t: • TT 
 
 with hunger", "emaciated". Cf. 2V~) ">1C Deut. 32: 24. The 
 
 t t •• : 
 
 sense thus gotten from HIQ is especially suited to the context, 
 
 TT 
 
 the contrast here being between "the best of the sheep, and of 
 the oxen, and of the fatlings, and the lambs, and all that was 
 good" on the one hand, and that which was lean and poor on the 
 other. See further the Introduction, p. 18. 
 
 18. Dnk Dni^S ~IU "Until they be consumed", R. V. The 
 
 exact translation of the present text would be "until they con- 
 sume them", which makes no sense here. B, L, and Syriac have 
 DHN ^ni>>5 ~iy, while Vulgate renders DHl^ IV- Our 
 present Hebrew text is a combination of these two readings. 
 With DHfc ^nfe IV cf. Jer. 9: 15; 49: 37, and for Qn1^3 IV 
 cf. .1 Kings 22 : 11 ; Ps. 18 : 38. See the Introduction, p7l7. 
 
 CHAPTER XVI. 
 
 16 - |y? y*] 1 P^ " A man who is a skillful player on the 
 harp", R. V. In no other instance is V"V followed by another 
 participle denoting that in which one is skillful. The regular 
 construction is V"V followed by inf. But the piel ptc. of the 
 verb designating the profession is often used of one who 
 habitually or skillfully is engaged in that profession. The two 
 readings j^p tfifc "a player" and jaj y-p J2'\S$ "one skillful 
 in playing" are both suitable here, and the Hebrew is a combina- 
 tion of the two. Cf. v. 18. LXX have w8 P a dSSra ^dxxav, the 
 second of these two readings. 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 35 
 
 CHAPTER XVII. 
 
 13. bvttf nrm 0S7 D^ian ^ \j? nt^D' d^i_ "And 
 
 the three eldest sons of Jesse had gone after Saul (to the battle) ", 
 R. V. One of the two verbs in the Hebrew is superfluous (see 
 Driver, ad loc.) . A and L have *u iiroptvO^av at the beginning 
 of the verse, A then repeating iiropevdrjirav in the place of l^vi""!- 
 
 : t 
 
 Syriac has only the first verb. It is quite plain that two readings 
 have here been combined. In one of them the verb was O/"!- 
 in the other OTTl The rest of the sentence was in both cases 
 
 : t 
 
 the same. If we keep the first verb and delete the second, we 
 have a regular consecutive clause, carrying forward a step the 
 action of the main story. Granting the second verb its place in 
 preference to the first would render this clause a link in the 
 chain of circumstantial clauses which, precede and follow it. It 
 is well to note here that each verb is exactly in its correct place 
 in the sentence, wherefore no attempt could be made to actually 
 combine the words, as in other places a perf. and an impf. quite 
 often are combined. Cf. Deut. 33 : 16 ; Job 22 : 21. Whatever 
 else of the two words could have been combined, their positions 
 could not. 
 
 CHAPTER XVIII. 
 
 i. WDtt fnjlrp nnx"! in tfe:3 mtfpj fralir tfwi 
 
 : - : Itt : •• t ■-•:-.— ■ t v : t: I: • I t t : -.-•.■ : 
 
 "And the soul of Jonathan was knit with the soul of David, and 
 Jonathan loved him as his own soul", R. V. The Ketib has been 
 pointed farWl by Kittel and Driver (ad loc), and in support 
 
 t •.•:■.— 
 
 of this form of the suffix the latter cites Hos. 8 : 3 lDTT ; Ps. 
 35: 8 nri^n; Jer. 23: 6 ftnpi ; Eccl - 4: 12 iCpn\ Josh. 2: 4 
 
 UDJfni • K is interesting to note, however, that all these pas- 
 sages, with the exception perhaps of Ps. 35 : 8, bear the marks 
 of combined readings. In Hos. 8: 3 lCll" 1 could very well be 
 pointed as a plur. ?|©"n^"' (so the Greek) , since its subject ^lK 
 in sense is plur. 1JOT m Jer. 23 : 6 is considered by Ges. a 
 
36 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 "forma mixta" (74 e ), i. e. it could also be considered a plural (so 
 the Syriac) with an indefinite subject. In Eccl. 4 : 12 i£prP D&0 
 ill? HDJ?" 1 D^'H "in^n, generally translated "And if a man 
 prevail against him that is alone, two shall withstand him", the 
 uncertainty of the reading is obvious, for the suffix is useless ; 
 neither Grk. nor Syr. had it in their text, but took as the subject 
 of the verb "inNi"!. which naturally means "the one" of the two 
 companions, as in vss. 9 f . The text in Josh. 2 : 4 is known to 
 be corrupt (Driver, Notes on Samuel, I, 18: 1), but the word 
 found in the Hebrew seems to be clearly a combination of DJSV^l 
 "and she hid them" and US VI " and the y hid " (intr.) . To this 
 evidence should be added )ity) in 21: 14 and ^ in II S. 14: 
 
 6, on which see notes. 
 
 A form l2n&01 is therefore extremely unlikely here, in any 
 text. The plur. of the imperf. is not to be thought of, and the 
 perf . with suffix would be out of place. The imperf. sing, without 
 suffix, however, would be most natural here; and this, DH^Ij. 
 
 was probably one of the two readings, THJin" 1 being taken as the 
 
 object. What M.T. gives us is the only natural combination of 
 this latter form and the Qere. 
 
 CHAPTER XIX. 
 
 20. DrP/1? "2^ "ipy ^NID!?^ "And Samuel standing as 
 
 one appointed over them", Driver. "Both ptcpp. are represented 
 in LXX, but the combination is peculiar and suspicious" (Driver, 
 ad loc.) Syriac has only >:Lc . This is probably another instance 
 of two readings preserved in full in the text. Either one of the 
 participles would be sufficient for the meaning. 
 
 CHAPTER XX. 
 
 2. -Q1 ^N nfcV° )b- Ketib r\ty )b "If my father had done 
 
 T T • T V -:- T T 
 
 anything", etc. (the conclusion is left out, or rather implied). 
 Qere n^S?" 1 &6 "My father doeth nothing" etc. Either con- 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 37 
 
 struction is idiomatic and plausible. For analogy to K. cf. Gen. 
 50: 15, where y) introduces a conditional clause without an 
 apodosis. 
 
 24. Urhn hv Tb^n 3£^1 "The king sat him down to eat 
 
 food", R. V. Ketib ?]); Qere 7^. 7^ and 7l? are very often 
 
 interchanged, seemingly with the same meaning. See the note 
 on 14: 10 and 12. 
 
 25. TpH 2PlD ^"W1D by-rfcftPi ypH "And the king 
 sat upon his seat (as at other times), even upon the seat by the 
 wall". 7^ and 7I? are both preserved in the text, as indication 
 
 of two traditions. Cf. preceding note, where both are combined 
 in one word, and have a marginal note. 
 
 30. itfi pb nfiN mb "G wt tibn "Do not i know 
 
 - • I v : t - •• • • : -t -: 
 
 that thou hast chosen the son of Jesse?" R. V. LXX have 
 ixiroxoi for Hebrew "*ir!3, i- e. they read "1211 • Syriac has 
 ^^] ? oi^iis aj] log. After "IP! 3 m the Hebrew we should cer- 
 tainly expect 3, for there is no instance of 1113 being construed 
 with 7. The 7 . therefore, points out to the reader that two 
 readings are contained here: one i$*> J3 1 ? HV)# ""QI1 "'3, the 
 other i$i ]33 |~|PlN "IPD "O- If one consider what an easy 
 mental process the transposition of letters is in rapid writing, 
 it will be readily understood how effective a combination of this 
 kind would be. 
 
 CHAPTER XXI. 
 
 3. i±bx ^S QipP % ^tHi" 1 CHinn m\ "And I have 
 appointed the young men to such and such a place", R. V. The 
 form "Tl}HV has been explained by Ewald (Lehrbuch der 
 
 Hebraischen Spracke, §125 a ) as the Po'el from y"p . But this 
 
 — r 
 
 is found nowhere else. The phrase is clearly a combination of 
 two readings: one "B DlpD b# THST 1 . the other Tiyiln 
 
38 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 " 3 DlpD nX- The first reading is suggested by the prep. ?N 
 and *> as the first consonant in "'niHi' 1 ; the second reading has 
 its characteristic marks in the verb ,, Pljn1\ ^ being the only- 
 letter that requires any change. Driver (ad loc.) suggests either 
 the Po'el or the Hiph. of "l^T as the correct reading here. The 
 
 — r 
 
 form of the combination, however, does not suggest the Hiph., 
 and the Po'el is found nowhere else. Of course the prep. ?N 
 cannot be retained with the Hiph. of J7"p, but must be changed 
 to n^; the form TllHln taking two accusatives. Either of 
 
 these two constructions fits the context, and is grammatically 
 correct. 
 
 9. rVJn T]T nnn !"IB & 1 p^l "And is there not here 
 under thy hand a spear"? This unparallelled pointing of p^t 
 directs the reader to look for the reason for it. Inevitably an- 
 other short word pointed with Hireq comes to mind, namely Q^. 
 Objection to reading Q^ here has been raised (Driver, ad loc.) 
 on the ground that Q^ is always used to introduce a question 
 to which the answer "no" is expected, and therefore is not suit- 
 able here. But granting that such is the exclusive use of Q^ 
 in questions (except where it continues the force of ~ ) , its use 
 
 here would still only emphasize the contingency in the question : 
 "there is not possibly here under thy hand a spear?" The Greek 
 also supports this reading ; it has "iSe el Ivnv ivravda vtt6 ttjv x&p°- 
 o-ou S6 P v rj p fxcj>aia. On the other hand, the combination $"> p^ 
 
 is found also in Ps. 135 : 17, and seems to be recognized as legiti- 
 mate (BDB, p. 442 a ) . The question, if that reading be adopted, 
 would be indicated in the tone of the voice. The text clearly 
 contains two readings, the one being J-|B ttf* 1 DN, as supported 
 by the Greek; the other may be either HB D'" 1 TN, with the 
 parallel in Ps. 135: 17 as its justification, or else may be only 
 ("IB r&$> which the Syriac seems to represent, U>'d &*!». 
 14. DiT^lD l^yip n« 13EH "And he changed it, even his 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 39 
 
 behavior, before them", Driver. "The emphatic anticipation of 
 an object such as DJ7CD is not probable, and the form of the suffix 
 
 — rare even in strong verbs — is found only once besides with 
 a verb n"7, II S. 14: 6, where there are independent grounds 
 for questioning its correctness. No doubt IJIE^I is an error of 
 
 transcription for n^^V (Driver, ad loc .). But on the analogy 
 
 of several other passages (see the note on 18: 1) including II S. 
 14 : 6, where Driver admits that the consonants were intended to 
 designate a plur. form, it seems- at least plausible that the conso- 
 nants here were meant to suggest a 3rd masc. plur. of the pi'el 
 impf. U^T, "and they changed". Note that both Grk. and Syr. 
 
 read singular suffix, "pj 1 ^, "before him" (Achish) . The peculiar 
 form is here, as well as in the similar passage referred to, meant 
 to call the reader's attention to the two traditions, one having 
 the sing, (in this case W^ or rG^I), the other the plur. (in 
 
 this case ^E^D- Pointing the "] as 1 seems to be a conventional 
 
 way of indicating that there is possible a choice between sing, 
 and plur. Cf. the notes on 18: 1 and II S. 14: 6. 
 
 CHAPTER XXII. 
 
 13. hWC "6n "lOtf"! "And Saul said unto him". The Qere 
 is vh#- This case and its only parallel Zecb. 2: 8 (cf. y)y and 
 
 T " T T 
 
 the notes on 2 : 10 and II S. 20 : 8) have been associated by Kbnig 
 (II, p. 303) with the plural nouns which have the 3rd masc. sing, 
 suffix ) attached without an intervening ">. But the situation 
 here is different. Both here and in Zech. 2: 8 y)H follows im- 
 mediately upon ID^T; And considering that ""|ft&< is followed 
 innumerable times now by p, now by 7N, there can be no doubt 
 that I7N is a combination of y) and T;N- 
 
40 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 CHAPTER XXV. 
 
 14. D'HI^hD in^ ""ljn "One of the young men". This ex- 
 pression is certainly too full. Greek B represents only LPiN 
 D^UJinD; L has «V tu>v TTo.i%a.piu>v airo tu>v -n-aiSwv ; and Syriac agrees 
 
 • t : - 
 
 with the Hebrew. It seems that two readings must have been 
 combined here, one being D'Hl^HD IH^. cf. 9: 3; the other 
 
 •t : - •• t v 
 
 was perhaps "infr? "IW " a certain young man". 
 
 TV 
 
 17. irP3 h3 bv) U^IN ^ "Against our master, and 
 against all his house". ^N and ?J? are here interchanged in the 
 same expression, each one representing a different tradition. 
 See Introduction, p. 19, and the note on 14: 10 and 12. 
 
 34. ^nNlpb Tl&Oni mnD "6^ "Except thou hadst hasted 
 
 ti: • : t - : : - • •• . 
 
 and come to me". This form, Tl^Om* has already been pro- 
 
 : t~ 
 
 nounced a combination by Nestle (ZAW. XIV, 319) . There is 
 seemingly no difficulty in accepting it as a combination of the 
 impf. i&QPVI an d the perf. nfcOt. But the Massoretes give a 
 
 • . t - : t 
 
 Qere r\^"2V)V- This seems to be a kind of correction of the Ketib 
 TlfrOm ; see the Introduction, pp. 22 ff. For other explanations 
 
 : T - 
 
 of this and similar cases see Ges. 76\ 48 d ; Kon. I, p. 647. 
 
 CHAPTER XXVI. 
 
 12. ^W ^nfcf&np "From Saul's head", R. V. These words 
 are admittedly suspicious (see Driver, ad loc.) . One question is, 
 can "SOD for "iOp?2> be considered permissible? Another 
 arises concerning this lone instance of 1 attached to the constr. 
 state of the fem. plur. before an independent word (otherwise 
 only occurring before suffixes) . And independently of these dif- 
 ficulties comes the objection, that HE^OD is nowhere else found 
 except with a suffix. With the aid of the Greek, which renders 
 TnitfiOpp, the problem is solved if we assume that two read- 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OP SAMUEL. 41 
 
 ings have been combined : bw& £'&OD "from Saul's head" and 
 
 T 
 
 TnttfiTlDD (lit-) "from the parts about his head". 
 
 t -: - : • 
 
 15. Tpj'-IN 7N filCtf N7 nub) "Why hast thou not kept 
 
 ! v -: v t : - t t - : . 
 
 watch over thy lord?" *)fi$ is not generally used with 7^. The 
 next verse has 72? in the same kind of construction. The two 
 cases are found in close enough proximity to suggest that two 
 traditions existed, one with 7^ in both places, one with ~>V- 
 Cf . note on 14 : 10 and 12. See also II S. 11 : 16. 
 
 i6. w'*np -lEto D?sn nns^ m\ rfcnn rron ^ n«i 
 
 "See where the king's spear is, and the cruse of water that was 
 at his head". We should expect here that the construction would 
 be the same for the two objects mentioned — either ifc$ in both 
 
 places, or J"lfr$. The fact that we have in one place *>&$, and in 
 
 the other riK is probably a clue to alternative readings, one with 
 
 ">K uniformly, and one with n& Greek B probably had 1^ and 
 
 H^ in reversed order, for it reads : iSe Si) to 86pv tov /WiAe'cos ko1 6 
 
 <j>aKos tov vSaros, wov io-TLv to. 7rpos Kt^aX^s avrov, L seems to represent 
 ">&$ in both places by the following translation : 181 wov Io-tI t6 86pv 
 
 TOV fio-CFlXitilS Kol 6 <£(XKOS TOV V&O.TOS TO, 7TpOS K€Cpa.Xrj'S O.VTOV. 
 
 CHAPTER XXVII. 
 
 10. DVT! DHiD^B bit "Against whom have ye made a raid to- 
 day?" R. V. This translation is more than can be gotten from 
 the Hebrew. It is made on the basis of the Greek and the Vulg., 
 which represent ip 7^. Driver prefers to read jN, "where" 
 
 or "whither"? But it is more natural to suppose that the word 
 ">£ had fallen out, as indicated by LXX and Vulg. The latter 
 
 part of the verse bears out this supposition. In David's answer 
 to the question three districts are mentioned, against which he 
 had made a raid. But 72? is use d twice and 7N once for this 
 
 Library Publications. 3. 
 
42 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 "against". This circumstance makes quite certain the conjecture 
 that ^ is a combination of ?]] and y#. Compare this situation 
 with that of 20: 24, 25. The two seem almost identical. In 
 20: 24 ^ and ^ are combined into bV- Then follow in v. 25 
 one 7j? and one ^ . Here the combination is made differently 
 — ^ instead of py — and no marginal note is found. But the 
 subsequent interchange of 'jy and 7^ was perhaps considered 
 sufficient explanation of 7^. 
 
 CHAPTER XXVIII. 
 
 2. inn HPlN "Thou shalt know". Greek and Vulg. have 
 
 inn nni? "now thou shalt know". In several places where 
 _ .. T _ 
 
 nnN or nny occurs, there seems to be evidence that two tradi- 
 
 T - T _ 
 
 tions existed, one with nnN> the other with nnU- No attempt, 
 
 however, has been found in the books of Samuel to combine the 
 two in the text, and no marginal note throughout these books 
 directs us to read one of these two words instead of the other, 
 or rather as an alternative reading with the other. It is probable 
 that the similarity in sound between the two words was con- 
 sidered sufficiently plain suggestion to the reader that two read- 
 ings were possible, in cases such as are referred to above. Cf. 
 v. 22 ; II S. 7 : 20 ; 18 : 3. It may be said in regard to this passage 
 that to have the personal pronoun expressed as here is an unusual 
 and emphatic construction, which would readily suggest the 
 easier and more usual nni?; while, if t~iDV were found in the 
 text, it would not so readily suggest the more unusual PinK- 
 15. Tj'p nfcnpNT. "Therefore I have called thee", R. V. This 
 
 has already by Nestle (Marginalien, p. 15) been explained as a 
 double reading. In reference to the many different attempts to 
 account for the word he writes : "1st es so schwer zu sehen, dass 
 die Schreibung uns zur Wahl stellt, ob wir an Nip rufen oder 
 
 an i"Hp begegnen denken wollen: so rief ich Dich fcODSO, so 
 
 kam ich Dir entgegen d. h. zu Dir mpNI?" 
 
ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 43 
 
 CHAPTER XXX. 
 
 21. DTitf; 1 ] "And they made them to abide". The defective 
 writing of Q^^l suggests the sing. D^K^l.. though not neces- 
 sarily. In this case, however, the sing, is borne out by the Ver- 
 sions, and the writing Q^"!^'^ instead of D^E^I may here be 
 
 intended as a combination. The context is suited by either 
 reading. 
 
 22. ]nj $h "•pi? O^n tib *ltfN "Because they went not with 
 us, we will not give them, etc.". ">ftj? refers to "the group re- 
 garded as a unity, and spoken of accordingly in first ps. sing." 
 (Driver, ad toe). Nevertheless, the juxtaposition of ifty and 
 
 THJ is an indication of two readings combined, one with the 
 plural uniformly, and one with the singular. The Versions read 
 plur. )2t2V- Cf • II S. 21 : 4, where the situation seems to be the 
 
 T ■ 
 
 same, but where the Massoretes evidently considered a marginal 
 note needed as an explanation. The changing moods of a scribe 
 may account for the lack of a Qere here, while II S. 21 : 4 has one. 
 In a certain mood he would consider the combination obvious 
 enough without a marginal note, in another not. 
 
 CHAPTER XXXI. 
 
 ifo^b h"dD DTlE^S "And they cut off his head, and stripped 
 
 ■■ - : • t ■ : • : 
 
 off his armor, and sent into the land of the Philistines round 
 about, to carry the tidings", R. V. The critical word here is 
 in 7^1 • It represents two readings: ! irPl£' , ' ! '] "and they sent 
 (word)" and 'inpS^l "and they sent (the objects mentioned)". 
 The first reading is suggested by the context, since no object is 
 expressed and *1!£07 follows. The second reading is actually 
 found in the text unmistakably supported by the pointing. The 
 object, expected after the Pi'el of r\/V> can easily be supplied 
 
44 ALTERNATIVE READINGS IN THE FIRST BOOK OF SAMUEL. 
 
 from the clause immediately preceding. Greek L represents the 
 first reading, and B and the Syriac the second. It should be 
 noted, that if the word had been pointed as Qal ^rP^I. there 
 
 would have been nothing to indicate a different tradition. This 
 was evidently the only possible way to combine the two readings. 
 See Introduction, page 21. 
 
 9. DVn m) DiT^ rP3 ")feO^ "To carry the tidings 
 unto the house of their idols and to the people". Greek and 
 I Chr. 10: 8 have flN instead of JT3- Syriac has an equivalent 
 
 to the Hebrew except for reversed order. It is possible that two 
 readings have been combined here ; one identical with our Hebrew 
 text, the other one having J"lNl instead of rp2» as I Chr. and 
 
 the Greek. Only the one reading is put into the text, the other 
 one is suggested by the context and by QJjn J"lNl . It is a case 
 
 t t :■ : 
 
 similar to the one described in the preceding note. See Introduc- 
 tion, page 21. 
 
Alternative Readings in the Second Book 
 
 01 Samuel. 
 
 CHAPTER I. 
 
 21. 1D7 -Q " 1 *in "Ye mountains of Gilboa". The mixture here 
 
 - : ■ - "T 
 
 of the construct state with the preposition is similar to that in 
 I S. 9 : 3 and II S. 10 : 9. 
 
 CHAPTER II. 
 
 9. "bi?V'byV' t PyV'^V'^V'^ in?^ "And he made him 
 king over", etc. This is perhaps the most striking case of ")^ 
 and py interchanged, 7^ thrice, then 7}? thrice. Compare it 
 with I S. 9 : 4, where the sing, and the plur. are mixed about as 
 evenly as here 7}7 and 7^. 
 
 CHAPTER III. 
 
 13. TlN^n ^£37" DN "O ^S n« nmn tf? "Thou shalt not 
 l_. ....... . . . _ T .. ... 
 
 see my face, except thou first bring", etc. "0D7 is not used as 
 an adverb elsewhere, and cannot be translated "first" legitimately. 
 Greek and Syriac do not support "0D7- "DK "O and \3D7 are 
 mutually exclusive" (BDB, p. 474 b ). We have here plainly a 
 combination of the two readings: JlfcTDn QN "*3 "unless you 
 
 bring" and TJijTDn "^C? "before you bring". 
 
 is. n« Win nay nn T3 nb*6 -rn ^ nDN rrtrp 
 
46 ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 
 
 ^"I^' 1 ifiV "Jehovah hath spoken of (or to) David, saying, 
 By the hand of my servant David / will save my people Israel". 
 This translation, requiring the change of JTEfin into ITttfiN, is 
 supported by the Versions. But cannot the text be under- 
 stood in another way? If we read U^'in, Hiph. inf. absol., or 
 y^'in^. inf. constr., we may translate: "By the hand of my 
 servant David (shall come) the salvation of my people Israel." 
 That both of these possibilities are intended by the present text 
 seems very probable. 
 
 25. 7jN"Qb nW ?]NtflD 21N "Thy going out and thy coming 
 in". Keiib TI^I^D; Qere TjipiD- The Qere of this verse and 
 
 the same word (except for the suffix) found in the text in Ez. 
 43 : 11 have been perplexing. Driver (ad loc.) says : "Why the 
 abnormal (and incorrect) form TjNDlD should be substituted as 
 
 Qri unless for the sake of the assonance with TjNUlD. is not 
 
 apparent." Konig II, p. 127, "nur Nachahmung des vorausge- 
 henden Wortes." But these explanations are not wholly satis- 
 fying. And there is another explanation which is much more 
 natural, and is borne out by evidence. The reason for substi- 
 tuting TjfrOlD as the Qere was to indicate to the reader that 
 
 some authorities have TJ^lfiO HN and Tj^'DP n^ in reversed 
 
 order. Greek L has these two words in reversed order : rqv ao-oSoV 
 (tov koI rfv Z£o86v (tov . If the text had been left as it was — correct 
 enough in itself — this tradition with the words in different order 
 would have been lost. It is not to be supposed for a moment, 
 then, that anyone actually read Q. as given here in the margin. 
 The note was there simply to point out the double reading. The 
 case in Ez. 43 : 11 is exactly the same, except that there the text 
 itself already had the double reading preserved, and no marginal 
 note was considered necessary. The Syriac of the Ezekiel pas- 
 sage represents the tradition that has the reversed word-order. 
 This Qere and the one in I S. 25: 34 should be noted as most 
 remarkable instances of the Massoretes "improving" the text, in 
 the one case to clarify the combination of two readings, in the 
 other to bring about such a combination. See the Introduction, 
 page 24 f. 
 
ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 47 
 
 29. TDK rV2 73 bm 3^ &tih by 17IT "Let it fall upon 
 
 • t ■• x v : t : t 
 
 the head of Joab, and upon all his father's house", R. V. Inter- 
 change of 7N and 7J7, denoting double tradition. See the note 
 on 2 : 9. 
 
 CHAPTER V. 
 
 2. bvnvr nt* ^sm wsto nrvn nnN "it was thou that 
 
 •• t : • v ■ •• - : • - t ■ t t - 
 
 leddest out and broughtest in Israel". Ketlb 13^ JOlflD rPTl 
 
 T • T 
 
 "Thou wast leading out and bringing in," or "Thou wast wont 
 
 to lead out and to bring in (Israel)." Qere ^3B!"|1 N^lEi"! rPH 
 
 . .. - . . - T . T 
 
 "Thou wast the one that leddest out and broughtest in." Here 
 
 it is especially the division of the words that indicates to the 
 
 reader the two constructions. K. makes rVn and the following 
 
 T " T 
 
 participles without the article a sort of compound tense, denoting 
 customary action in past time (Ges. 116 r ), while in Q. the part- 
 iciples are the predicate after rVTl- It is worth noting, that 
 
 T • T 
 
 there was seemingly no way of showing directly in the text in 
 regard to the article prefixed to ">3D that it could be kept in the- 
 
 text or left out, as was the case with the article of fc^lflD. 
 
 This H wa s attached to the preceding rTTI, the pronunciation 
 
 T ' T 
 
 and meaning of which were not thereby affected, and the reader 
 could be expected to treat the two articles uniformly, whether 
 he would keep them both or delete them both. See Intr., p. 14 f. 
 8. "ill ttfBJ m& "That are hated of David's soul", R. V. 
 Ketlb 1NJI2?; Qere ">&OE7. Of this verse Driver says (ad loc.) : 
 
 "It is easier on this passage to say what it does not mean than 
 what it does mean." Because of the evident corruption of the 
 passage it is hard to determine what the Ketlb is. But it may 
 be the 3rd plur. masc. perf. Qal ^J^. 
 
48 ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 
 
 CHAPTER VII. 
 
 20. ni?T nFlfcfl " For thou knowest." Here LXX read: 
 
 p t- 3| t : _ t T - : 
 
 koI vvv ah otSas. Syriac has : M "<-r- M. It may be that HDX 
 here was meant to suggest also i"ini7 ■ Cf. the note on I S. 28: 2. 
 
 T " 
 
 23. The mixture of 2nd and 3rd pers. and of sing, and plur. 
 in this verse has been explained by Geiger (Urschrift, p. 288) . 
 
 CHAPTER VIII. 
 
 14. D"0!tf Dfc DHN bl2 D"attJ DHN3 U&) "And he put 
 
 • • : t v : t : • ■ : v : v v t- 
 
 garrisons in Edom ; throughout all Edom put he garrisons." Greek 
 L agrees with the Hebrew, but B does not render D"Q^J Dfo- 
 Syriac seems to render only the second half of the phrase: 
 joo?1 oiNn^ |j^L-o ^o? >a J _oio. Two readings have probably been 
 put into the text here side by side: □"OyJ DHN3 DK^Ij and 
 D'OUJ DU D1"1N b'Dyi- One of the two is certainly super- 
 
 • • : t v: t : 
 
 fluous. 
 
 CHAPTER X. 
 
 9. ^fcnifr'O "H-lllS ^3P "Of a11 the choice men of Israel", 
 R. V. Ketib b&O^'G D'HIPD "The choice men in Israel"; 
 
 . _ •■ t: • : : 
 
 Qere ^fcOET 1 "HlPiS "The choice men of Israel". It would seem 
 
 that no marginal note would be needed to explain these words, 
 for the two readings are apparent even after a casual examina- 
 tion. Furthermore, the construct state before a prep, occurs 
 often in Hebrew and has analogies in other Semitic languages, 
 so that there are reasons for considering it grammatically cor- 
 rect (Ges. 130 a ). In the books of Samuel we have at least two 
 instances of this construction, I S. 9 : 3 and II S. 1 : 21, the notes 
 on which see. The purpose of the marginal note here seems to 
 be, therefore, not to explain the meaning of the phrase, but rather 
 
ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 49 
 
 to point out that some authorities read 7it*"}C1^3 D ,l- Vin3 and 
 others ^Iferi •'-flrQ- 
 
 t : 
 
 CHAPTER XII. 
 
 22. nln 1 "^JIT y-lV ">D "Who knoweth whether Jehovah 
 will not be gracious to me," R. V. Ketib "OJIT; Qere i}|ni- 
 
 "The correction of the Qri is unnecessary : the Kt. is exactly like 
 Joel 2 : 14 ; Jon. 3 : 9" (Driver, ad loc.) . But it is not a correction. 
 31. |3^S3 "In" or "with the brickkiln" or "brickmould." 
 Ketib p^D3; Qere |37ft2. K. is probably only an error; yet 
 the "2 and the peculiar context here suggest possible influence 
 of the phrase Tibtpb CtfN3 T2#n "to devote by fire to Melek", 
 
 instances of which are found in Jer. 32 : 35 and II Kings 23 : 10. 
 It should be noted too that in the preceding verse a word is found 
 which contains the radicals of Tj'Pft, the god, and by some (see 
 LXX) was actually read as QJDpD, Milcom. There is a possi- 
 bility therefore that K. here is intended to bring to mind the 
 phrase mentioned above, giving it the derisive turn, T3]?n 
 
 uibnb onfc- 
 
 : • : t 
 
 CHAPTER XIV. 
 
 6. ink nD'l -inNn nN -inNn 13*1 "But the one smote the 
 
 V T- T V T V T V T 
 
 other, and killed him," R. V. This is one of the passages cited, 
 by Ges. (60 d ) and Driver (on I S. 21: 14) as containing imper- 
 fects with the suffix !)j"|-> contracted to ). But see the note on 
 
 T 
 
 I S. 18: 1. In his note on the present passage, Driver admits 
 that "Probably the consonants "D^l were meant originally to 
 express a plural, which arose through a false interpretation of 
 "inNPrnN "in^n ( as though this were "one another") ; and 
 VnfcrnN in^n 71*1 should be restored from LXX." But the 
 change of "inNH fiN into VHN JIN* on the basis of rbv S.SeX(j>6v 
 
 T V TV • T 
 
50 ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 
 
 avrov being the original reading of LXX rather than rbv ha 
 (Driver, Notes on Samuel, p. 237, footnote) does in no way neces- 
 sitate the elimination of the 1 of 13*1. Compare Ex. 32: 27; 
 
 Mic. 7: 2; Gen. 11: 3, where the similar expressions E^N 
 
 VnN n^ and ^"117*^ ^ &**# are f °u nd with the P lur - of the 
 verb. It is rather the pointing of the ) as 1 that must be ac- 
 counted for. The situation here seems to be the same as in 
 I S. 18: 1 and 21: 14 (the notes on which see) as well as in 
 Hos. 8 : 3 ; Ps. 35 : 8 ; Jer. 23 : 6 ; Eccl. 4 : 12 ; Josh. 2 : 4, where 
 (with the exception, perhaps, of Ps. 35: 8) the ) could be pointed 
 >) , making the respective verbs plur. instead of sing, with the 
 suffix. The reason for the extraordinary pointing seems to have 
 been a desire to combine the sing, and the plur. into one form 
 which should give the reader his choice. In some other cases, 
 where evidently the same purpose was in the mind of the scribe, 
 he made a marginal note showing that there was authority both 
 for the sing, and the plur., e. g. I S. 12 : 10 ; 15 : 16. It seems 
 therefore that "in^H IlN "inNH l^l is a combination of two 
 
 T V T T V T 
 
 readings. One was Tn^H nN "iPIKn 13^1; the other was TT1 
 
 inxn na -irusn. 
 
 T V T V T V T 
 
 ii. D^n bx'z rvanno rprbx nirr na rbvn xi w 
 
 , T - " - . - .. | .. ... T . .. ).. .. - T T . . . 
 
 nnD7 "Let the king remember, I pray, Jehovah thy God, that 
 
 the avenger of blood destroy not any more." Thus, if we adopt 
 
 Qere n3*inp (intending ni3~inp? The vowel in the final 
 
 syllable may be merely a warning of the fact of a double reading ; 
 see the note on I Sam. 4: 13). What the Ketlb is, is a matter 
 of some uncertainty. Besides the context, the only clue to it is 
 the i inserted between 3 and pi- The consonants in the text 
 resemble the noun rP3"lD, "greatness" (cf. II Chr. 9:6). The 
 
 Syriac would perhaps seem to favor this word: IolX L.^? ^?2j 
 ~ ^ ^ " "-o V lie?? U?x-? li^aias, and the context would not be badly 
 suited by it. The translation, if this word be adopted as the 
 Ketib, would be : "Let the king remember, I pray, with (the help 
 of) Jehovah thy God, the greatness of the avenger of blood to 
 destroy", i. e., his great desire and power to destroy. The con- 
 
ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 51 
 
 sonantal difference between K. and Q. would then however in- 
 volve not only the \ as indicated by the text, but also the i~|, 
 of which the text makes no indication. Another possible Ketlb 
 is JVS *inp, which would make the translation : "Let the king 
 
 remember, I pray, with Jehovah thy God, (that) the house (i. e., 
 family) of the avenger of blood hasteneth-to destroy." The sense 
 thus rendered would fit the context admirably. We would have 
 here an urgent appeal that the king might waste no time in ful- 
 filling his promises of protection, since the avenger of blood, and 
 in fact "the whole family" (v. 7; note also the plur. WOtS^ 
 
 immediately following here in vs. 11) hastened to destroy. This 
 reading as the Ketlb would involve no more changes of the con- 
 sonant text than are already suggested by rVi^nD- 
 
 21. TPtyy "I have done." In some manuscripts this word 
 
 has against it a note, supplying rVi^I? as Qere. In others the 
 
 T / T 
 
 text has fVET and the margin TVEW- See Introduction, p. 12. 
 
 T • T • ■ T 
 
 The Versions favor TP£>J?- 
 
 CHAPTER XV. 
 
 8. " , JD'>£''' D"^" 1 DN "If Jehovah shall indeed bring me again 
 ..... T 
 
 (to Jerusalem) ", R. V. We have here a case, where neither the 
 
 Ketlb (hitherto thought to be ^ffl ; see Driver, ad loc, and 
 
 Baer) nor the Qere ^W' 1 seems to satisfy the demands of gram- 
 
 T 
 
 mar and sense. "Kt. ^E^ 1 'if he brings back, brings me back,' — 
 
 • T 
 
 an utterly un-Hebraic sentence. Qri ^fc^, from "2$** to dwell, 
 
 T - T 
 
 unsuitable beside ,, J2 1 L , ' , will bring back" (Driver, ad loc). But 
 can we not suppose that the Massoretes also knew that 21E'' 1 
 
 T 
 
 to dwell was unsuitable, ordinarily speaking, beside "O^E*" 1 will 
 
 bring back? Knowing that, they must have had some extra- 
 ordinary reason for putting "2)&' 1 m the margin as a variant of 
 
 T 
 
 ^W ', and nothing furnishes us with a more plausible reason 
 than the supposition that they wished by such pointing to indi- 
 
52 ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 
 
 cate that two readings were combined in this passage. They 
 knew that by pointing the first word 2lCtf\ the inf. absolute of 
 
 T 
 
 "2pi, they would inevitably lead the reader to continue with the 
 
 only logical form under the circumstances, namely ''J^E^'', "If 
 
 Jehovah shall indeed cause me to dwell in Jerusalem." This 
 reading is already assured, it should be noted, by the contrasted 
 "while I dwelt (TQI^D ) in Geshur," in the first half of this 
 
 very verse. The other reading is "OD" 1 ^' 1 "2WT{ QN "If Jehovah 
 
 ..... . T 
 
 shall indeed bring me again," with which the Versions agree. 
 With the form cf. infin. absol. ^3H • It is perhaps more prob- 
 able, however, that the "• is merely a 'warning letter' ; see note 
 on I Sam. 4 : 13. We find, then, that both constructions are about 
 equally well represented in the text, one word being pointed 
 entirely according to the one reading, the other entirely according 
 to the other. Since " , J^t£ , \ representing the verb ~2\£f), is so 
 
 similar to " l ,12" , D i '' from 3t^'\ it seems the Massoretes feared 
 .... - T 
 
 that the reading "i^ 11 ^" 1 S^d'H would be entirely lost, and so 
 ..... . T 
 
 wrote "> in D -1 ^''' instead of ), so as to suggest the root "2)$. 
 
 T 
 
 See Introduction, pages 22, 23. 
 21. DN "'3. Ketib QN ">3 "(by no means) but"; Qere 13 
 
 "surely". Driver here thinks "the Qri is right". But compare 
 his note on 13 : 33. The oath is sometimes used absolutely ( Jud. 
 8: 19; Ruth 3: 13; 1 S. 20: 21), and can be considered so used 
 here. The QX "13 then begins the statement in the sense indi- 
 cated above. The Qere is in form like 3:9; see Driver, ad loc. 
 
 CHAPTER XVI. 
 
 2. nnbrh)- Ketib nnbrb) "And for fighting" -,Qereun Vni 
 
 "And the bread". It is strange that the commentators have not 
 stumbled upon the meaning of the Ketib in this passage. Driver, 
 ad loc, describes it as "an example of the accidental repetition 
 of a letter from a preceding word." Baer gives the Ketib as 
 
ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 53 
 
 DPDPPl- The sense, if K. be adopted, is "the asses are for the 
 
 king's household to ride on (lit. for riding) and for fighting." 
 The Versions support the Qere. 
 
 13. ~)BV) " b}?p?l ^p'l Tfhn " rfcr\ ^Vrppl The mixture 
 of tenses in this verse is very exceptional, and may indicate that 
 different traditions existed. Because of certain freedom fre- 
 quently exercised in the use of tenses (cf. 13: 19; I S. 19: 23), 
 it is however a difficult matter to say definitely what these dif- 
 ferent traditions were. One may have had all the verbs after 
 TjlTTl in the infinitive absolute. The other probably had "IS5P1 
 instead of "lB^T, while otherwise agreeing with our present text. 
 
 CHAPTER XVII. 
 
 22. *ni?J #b "inN IV "Until not one was lacking." -|nN> 
 
 though not in the constr. state, is pointed with short vowel in 
 the ultima, besides in this passage, in Gen. 48: 22; Is. 27: 12; 
 Zech. 11 : 7. The reason for this is hard to assign. See Driver, 
 ad loc, and Ges. 96 (p. 282, Remark) . On the analogy of the 
 confusion between ""[fiN and "infr? in Gen. 22: 13 and elsewhere 
 one might venture the conjecture here, that "inN! IV was meant 
 
 to suggest the two readings "ll"!^ IV > with the translation given 
 
 above, and "inN IV in the sense "till after"; cf. Neh. 13: 19; 
 
 Lev. 14 : 43 ; Jer. 41 : 16 ; Job 42 : 7. The meaning of the verse 
 would not be changed by this rendering, since "iriNl IV is prac- 
 tically only a fuller form of "IV, and for "inN, defined by *"1^'N, 
 
 - t v v -: 
 
 another antecedent of the relative would be supplied in thought, 
 which would be equivalent to "inN- Another possibility is that 
 
 T V 
 
 one reading was *1£JN "II?. 
 
54 ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 
 
 CHAPTER XVIII. 
 
 3. d^n n"iE>y MhD nrw "o " For now are there ten 
 
 ■ t -: t t -: T t - ■ 
 
 thousand such as we." The sense of this statement in the present 
 context is not altogether satisfying. Vulgate and some editions 
 of LXX together with Symm. read finN for nFlU: "For thou 
 art the like of us (being) ten thousand," for thou art worth ten 
 thousand of us ; cf . Driver, ad loc. The case here is very probably 
 the same as in I S. 28 : 2, where nHN in the text suggested PlPltf, 
 
 T - T~ 
 
 which there was in accordance with the more usual construction ; 
 this time Hfli? is found in the text, and suggests by sound HfiN, 
 
 T - T 
 
 the more suitable of the two. 
 
 3. ""Wl? 1 ? T17D Mb iTPin "O "That thou be ready to succor 
 us out of the city," R. V. Thus if Qere "\)]vb is read, though 
 the article would be expected with ""PIMp. The Ketlb cannot very 
 well be ""p]y b, as generally given by commentators, for a Hiph. of 
 
 JH1 is doubtful (BDB, p. 740 b ). But because of the lack of the 
 article with the preceding word it seems not unlikely that T}?7 
 
 ' T 
 
 (originating in a mere blunder of copying) should be the Ketlb. 
 In that case the sense would be : "That we should have thee (as 
 aid) from city to city." But the reading of the Greek in this 
 place is significant; A, B, and L all represent *"P1J?7 by a noun > 
 and read *"pj?3 for "TJ7D. Whether by intention, therefore, or 
 
 " T • " 
 
 by accident, our present Hebrew text seems to be a mixture of 
 the two readings : ^vb TJ7D Mb rPPin and -^JD ^ rVnn 
 
 ■ • t ■ •• t v : ■ ■ t t v : • 
 
 ^liy^. It must be admitted that the two traditions do not become 
 apparent from the Hebrew alone, nor can it be said that the 
 marginal note helps a great deal toward making clear the com- 
 bination. Professor Torrey's suggestion that ""Hyb combined 
 *")1JP of the Greek reading and -)iyb as sequel to "l^D, is the 
 foundation for the solution offered above. 
 
ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 55 
 
 CHAPTER XIX. 
 
 32 - IT 1 " 3 ^ ^^V?? "To conduct him over the Jordan," 
 
 R. V. Ketlb HIT'S; Qere HlTi! FlN- As Driver correctly 
 
 points out (ad loc), we have here a mixture of two readings, as 
 indicated by K. and Q. above. Either construction is suitable, 
 and has parallel : K. in y. 19, Q. in w. 37, 40. Greek B represents 
 Q., while L has K. See Introduction, page 14. 
 
 43. )}h #&} nN£0 DN "Or hath he given us any gift," 
 
 T T ■ .... 
 
 R. V. The fact that the word riNC^.? has called forth two equally 
 
 plausible conjectures seems to be a good proof that the combina- 
 tion of the two readings contained in this word was well made. 
 Driver, in his note on this verse, suggests the reading fr^J 
 
 (inf. abs. Niph.) for nN£0; Gratz (G. d. Juden, I., p. 287;, fol- 
 lowed by H. P. Smith (ad loc), proposes to read nNfc'D "portion" 
 
 or "gift". Cf. Ges. 76 b and Kon. I., p. 632 f. There is nothing 
 in the combination for which the solution cannot be found in the 
 two readings proposed. 
 
 CHAPTER XX. 
 
 5. -in'"*! "And he tarried longer." Qere THi'l^Tn^l- The 
 Ketlb can hardly be anything else than ?rPl; cf. I S. 13 : 8. 
 There is no root TTP ( see Driver, ad loc.) . And how else should 
 a combination of the two forms TTl^l and /Pr" 1 ! have been 
 effected? 
 
 6. PIP nnK "Take thou." Greek B has «<u vvv o-i Aa/3£, A 
 
 I- T - 
 
 and L read koX vvv Xa/3e. Compare the notes on 18 : 3 ; I S. 28 : 2. 
 
56 ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 
 
 CHAPTER XXI. 
 
 4. ^ TN "It is not for us." Ketib "9 "for me"; Qere ^ 
 "for us". Q. has the plur. to correspond with the latter part of 
 the verse, where we read ^7 T^l- The whole speech of the 
 
 Gibeonites could have been put either in the sing, or plur. with 
 propriety. The sing, would then have been considered as repre- 
 senting the statement of an individual voicing the sentiment of 
 the mass. It is probable that in this verse some authorities had 
 ip |i{$ in both places, and others IJ7 p{<. In order to preserve 
 both these readings one ">? p^t and one )y} )*># were kept in 
 
 the text. The marginal note is not necessary here, since the text 
 itself makes clear the combination, but may have been put in 
 here by the scribe while he was in a mood of more scrupulous 
 conscientiousness than ordinarily. Cf. I S. 30: 22, where the 
 case is almost identical, but where the margin is silent. 
 
 9- 'inon DH1 "And they were put to death." Qere ^D HEill 
 with this translation: "And they died"; Ketib irO"! DHL trans- 
 lated as above. The double reading here evidently affects not 
 only Qni, but also ^HDH- Cf. the similar cases in 5: 2 and 21: 
 
 t •• : : : 
 
 12. The two forms Di~! and DSD seem to be "without appreci- 
 
 T " 
 
 able distinction in usage" (see further BDB, p. 241 a ). Greek 
 and Syriac read iriDH- 
 
 9. anjjfr Tap n>>nn o^fc^na map ^s "in the 
 
 • : 'I: - • : • ■ • t -It •• • 
 
 days of harvest, in the first days, at the beginning of barley 
 harvest," R. V. Ketib rhr\V); Qere ^11^2- 'T^nn is per- 
 fectly intelligible as the accusative of circumstance, without the 
 preposition which is prefixed by the Qre," H. P. Smith (ad loc). 
 Evidently both K. and Q. are defensible. But it seems that the 
 whole expression is too full, and it is probable that we have here 
 at least two different readings combined, or rather incorporated 
 in full. One seems to have been D" 1 J 5^X13 TUp " 1 D' 1 3, and 
 
 • t • It •• ■ ' 
 
ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 57 
 
 another Q^)'yp -)i\}p H^nriB; while a third may have been 
 D^fcftnn "l^p ^3 (see Driver, ad loc.) The Versions 
 
 • t -It ■• • 
 
 seem to support strongly the Hebrew text, except for the fact 
 that L does not represent D^&'iOD- 
 
 12- □ W'^Sn Dtf Q^n ~\p$ "Where the Philistines had 
 hanged them." Keith D^^Eil Q&*'; Qere D^H^S ~D£''- 
 "QTl£vC occurs much more frequently than D^nt^D": but the 
 latter is found (e. g:, 14: 7; 7: 13)," Driver, ad loc. Compare 
 the note on v. 9. 
 
 CHAPTER XXII. 
 
 28. ^2£n D^p"! hi? Try] "But thine eyes are upon the 
 haughty, that thou mayest bring them down," R. V. "The second 
 clause (of the verse) gives no suitable sense. For ""pjiyi 
 D^D~)~bv read niDl D^yi -Ps." H. P. Smith (ad loc). But 
 it is quite evident that two readings have here been combined; 
 one ^Sfc'F! D^DI "T^l "but the eyes of the haughty thou wilt 
 bring down", the other Q\2"l bv ^^V] "' 3U ^ thine eyes are 
 against the haughty". Greek L preserves the first of these 
 readings : «al o<£#aA./u.oi>s v^Xdv Tcoren/wrrtis. Syriac renders □"'J^Vl 
 ^S'^'n nlDI. like Ps. 18 : 28. 
 
 • : - t 
 
 43. DJJpIN! DjfHN "I crushed them, I spread them abroad." 
 
 This is evidently a doublet. The Versions have only one verb. 
 Ps. has DiT")N. Cf. H. P. Smith, ad loc. 
 
 CHAPTER XXIII. 
 
 8- i&b&n U$l "Chief of the captains," R. V. These two 
 
 words and the similar expressions in vv. 13, 18 are most illu- 
 minating and instructive as regards the methods employed by 
 
 Library Publications. I,. 
 
58 ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 
 
 the Massoretes in combining two readings. A word that suits 
 the context here excellently is the plural of Z*bvJ "officer" or 
 
 "captain" (cf. Ex. 14: 7; I Kings 9: 22; II Kings 7:2). It 
 follows naturally upon □"H^H in the earlier part of the verse, 
 and evidently must be taken together with the preceding word. 
 O^'V^n £''iO > then, "chief of the captains", is one of the alter- 
 native readings in the passage. But let the consonants too give 
 their testimony. Grammatically, it would be possible to add the Q 
 and read Qi&'^'n £'N"1 "chief of the thirty" (cf. Driver, ad 
 loc). The sense, too, would be satisfactory. But if one reading, 
 as already found, was □ 1 £ , 7&''n, and the other ^p7W>j, why 
 
 should not the □ have been written in the text? If the Q was 
 not put into the text, it must mean that at least in one reading 
 it should be left out. The natural conjecture is n&vu'n, re- 
 placing the " with ["!• That would give us n^7Wn &$h, 
 
 "chief of the three", grammatically correct and suitable to the 
 context. The three here referred to are of course those named 
 in vv. 8 — 12, Joshebbasshebbeth, Eleazar, and Shammah. When 
 all the circumstances are weighed, there can no longer be any 
 doubt that this is the second alternative reading. It should be 
 noted here that L has Trp&ros tZv rpiu>v. It is perhaps profitable 
 also to compare the situation in I Chr. 11 : 11 and 12: 19, which 
 are parallels to this passage. In both of these places the text 
 has □ , " , '1 L 5C£'n &i*h, and the margin directs us to read as one 
 choice D^& vtT'n u {O "chief of the captains". Note that here 
 
 the D » belonging to both readings, is put into the text. One 
 cannot help but ask, why the Massoretes placed a marginal note 
 against the passages in I Chronicles, while in II S. 23: 8 (a 
 similar combination 1 ) they did not. The answer may be this, that 
 D^'lpt'ij in I Chr., if pointed with the vowels of the word for 
 captains, could very easily be considered an example of the acci- 
 dental lengthening of "> to }. Thus, probably one of the two 
 traditions would have been lost, unless the margin had been 
 serving as safeguard. Cf. vv. 13, 18. 
 
ALTERNATIVE READINGS IN THE SECOND BOOK OF SAMUEL. 59 
 
 8. iriN DyS3 "At one time." Ketib inN; Qere nnN- 
 
 TV : TV T V 
 
 DyS is /em., and it is probable that the masc. form "inN 
 
 ■ T V 
 
 was intended to suggest as a variant for this expression the 
 phrase "IPIN DVG "in one day". 
 
 13. K^O D^'^'mE D^'V^ HTl "And three of the 
 thirty chief men went down," R.V. Ketib D" 1 ^'^'; Qere i~|£^£\ 
 
 • " t t : 
 
 In this case there can be no doubt whatever as to what K. is. It 
 must be D'Civt^', for it would make nonsense to say "thirty of 
 
 • • T 
 
 the thirty". The translation given above is that of Q.; K. would 
 be: "and certain captains of the thirty went down". Note that 
 though the same words are combined here as in v. 8, except for 
 the article, the combination is different. In v. 8 the consonants 
 stood for ~'l il )& and the vowels for □ 1 &vt£'. Here the case is 
 
 t : • ■ t 
 
 reversed. This cannot be considered simply an accident, for the 
 reason seems apparent. If this combination had stood in v. 8 
 with the Qere it has here, it would have been impossible to decide 
 whether the alternative reading should be D^v^'H or O^pk'n. 
 But here no such uncertainty need exist on account of the fol- 
 lowing word. Cf. the note on v. 8". B has *<" Karip-rjo-av Tpeis 
 
 'nrii toiv Tpia.KOVTa\ L kul Kariji-qcmv Tptls Ik toiv Tpiuiv ap^wv- SyriaC 
 
 lr=^ -^ U^lz cl-~Jc. 
 
 18. ^h'C~ VVTl "Chief of the three", R. V. Ketib COtf" 
 "the thirty" ; Qere r\&b&~ "the three". Cf. the situation here 
 
 T - ~~ 
 
 with vv. 8, 13. We should expect here the Q to be written in 
 the text at the end of the word (see rule 4, p. 22) . The reason 
 it is excluded from the text seems to be the fact that the vowels 
 of the two readings are so similar that if the consonants were 
 
 allowed to stand for Q">c6^'n alone > tne vowels could not be 
 expected to indicate the alternative, nc'^&'n . clearly enough. 
 Now, part of the vowels and part of the consonants stand for 
 each reading. Evidently rule 2, Introduction, p. 22, has had to 
 vield in this case to the rule of clearness. 
 
60 ALTERNATIVE READINGS IN THE SECOND BOOK OP SAMUEL. 
 
 CHAPTER XXIV 
 
 4 ^nn ^& ^V) 2N1 1 "?N T]!?sn W plIT.l "But the 
 king's word prevailed against Joab, and against the captains 
 of the host." Two traditions are here preserved, one with ^ 
 throughout, and one with jty. Cf. 2 : 9. 
 
 13. ?!D"n Nl~l rn")^ ijcb "Before thy foes while they 
 pursue thee", R. V. This verse combines the sing, and the plur. 
 in a way similar to that in I S. 30 : 22 and II S. 21 : 4. Cf . also 
 I S. 24: 5; 26: 8. Syriac has sing, in both places, while Greek 
 has plur. uniformly. One tradition must have been 7P' m ])i "JO? 
 TfC-n DH1, the other TjDTl Will TpS ">JC^. 
 
>*h> 
 
 **' **.,, 
 
 fa - 
 
 #**& 
 
 *> 
 
 ArtHkisczw