Studies in Hebrew Roots and Their Etymology Author(s): Joseph Reider Reviewed work(s): Source: The American Journal

of Semitic Languages and Literatures, Vol. 34, No. 1 (Oct., 1917), pp. 66-70 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/528515 . Accessed: 28/01/2013 17:52
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to The American Journal of Semitic Languages and Literatures.

http://www.jstor.org

This content downloaded on Mon, 28 Jan 2013 17:52:28 PM All use subject to JSTOR Terms and Conditions

66

THE AMERICAN JOURNAL OF SEMITIC LANGUAGES

70. GIMMA KI-AG SIS-UNU-KI = " 71. " " " " DUG-GAR-KI = " 72. " " " " KIA-KI = " = " " " VI.1. " " KI(-KI

-ri " " " Ki-e-hi " Ki-vi

Meissner, SAI, 6168, 8350, has exploited some of these results as he did one variant of Col. VI.2 in SAI, 4035. With respect to the sixth column it must be noted that A crowds 92 entries into this column, while B has only 60 lines. No two copies arranged their entries alike in columns. But they maintained the same order of entries. The nearest conclusion is that the whole text registered in all 72+72+72+72+72+92 entries, or 452. Many duplicates show the or ends of lines alongside those portions sufficiently preserved beginnings to be useful. Had the lengths of columns always been the same, these traces might have helped to fix the place of more fragments. But, to take one example only, Col. 11.5 is on the same level as Col. 1.5 on A, while Col. II.49 is on the same level as Col. 1.45 on B. Delitzsch gave a complete column down to line 38 (entry 44) and then from line 50 (entry 55) to line 75 (entry 92). We can complete from duplicates all the ideograms, but can give no Semitic equivalents. For the whole text the duplicates give 83 new entries and complete 28 of those already partly known, thus adding about one-fifth. Some day I may have the opportunity to publish the fragments in full. It does not seem worth while here to record mere variants nor set down unexplained ideograms. C. H. W. JOHNS
CAMBRIDGE ST. CATHERINE'S COLLEGE,

STUDIES IN HEBREW ROOTS AND THEIR ETYMOLOGY I The three roots p , ppi, and pbu were hardly distinguished in post-biblical and talmudic times when confusion of roots was the outstanding ben Saruk, the champion of the feature in biblical exegesis.' Menah.em biliteral (and also uniliteral) theory of Hebrew roots, throws them all together under pi 2; and even IHayyuj, the iconoclast of the biliteral fad, still con. with those whose origin is fuses words whose origin is Ibn p5
1 On violations in the Septuagint cf. Frankel, Vorstudien zu der Septuaginta, p. 200. Confusion of roots is also rampant in the minor Greek translations of Aquila, Symmachus, and Theodotion, as pointed out in my " Prolegomena to a Greek-Hebrew and HebrewGreek Index to Aquila," JQR (New Series), IV, 578 if. As to Talmud and Midrash cf.
Die Anfdnge der hebrdischen Grammatik, Bacher, hebrdischen Sprache und Schrift, pp. 69 ff. p. 6; see also Gesenius, Geschichte der

2 M It trT~ln,
3

ed. Filipowski, p. 179.

Verbs in Hebrew, ed. Jastrow, p. 268 of the Arabic text.

Evenp!

The Weak and Geminative

inthisgroup! Job31:27is included

This content downloaded on Mon, 28 Jan 2013 17:52:28 PM All use subject to JSTOR Terms and Conditions

CRITICAL NOTES

67

is the first to distinguishthe three,'and he is followedby who Ganahl in turn is followed by all modern lexicographers.3But, while KimIh.i,2 this root classificationwas effected, the etymology and consequentlythe meaningof these wordsremainedobscure. Thuispj was persistentlycomparedwith = Arab. "drive,impel,urge,"fromwhich is derived U "leg,"4 pit~ 3 and Brockelmann' until Barth5 suggested "long, desire,"whichcovers .J.LA .' As to pp~, associated completelythe Hebr. alreadyIbn Ganalh ;l it with Arab. ? in the sense of "walkingto and from, perambulating,"8 and this has been adopted by all lexicographers, both old and new, who in additioncompare "leg," hence"walking."' Finally,p3t is or else madeto correspond eitherleft withoutan ..= to VI.2:= 4 etymology'0 in the sense "to possess."" The purposeof this article is to subject these roots to a new investigation throughan examinationof the passagesin which they occur. Starting with Prov. 28: 15, :=i? 0~ "it, we see at a glance that pp?. pp.l,in being parallel to =,J , must signify something akin to it, something the nature of emitting a sound, and not, as is generallyassumed,"walking to and from." Now, if we consultthe Arabiclexicon,we findthat a , the adduces s.v., besides many other significavery same root which Ibn Ganahl tions, also has the meaning"to spreadin the sky (of lightning),"12 henceto
I Kitdb atl-UziU, ed. Neubauer, s.vv. 2 n'JTgj* "f)"Z , ed. Biesenthal and Lebrecht, s.vv. Kdnig, etc. 3Cf., e.g., Gesenius-Buhl, Brown-Driver-Briggs, and Gesenius-Buhls5, s.v. 4 Notably Brown-Driver-Briggs 5 Etymologische Studien, p. 46. 6 Grundriss der vergleichenden Grammatik, I, 167.

7We

must not be too scrupulous with reference to the irregular equation words as

especially in the face of such old and well-established

= , etc. I ut ?= the accepted rule will arise in the case of words with sibilants (especially in triliteral roots with two sibilants); similarly Vollers (ZFA, IX, 197, 209) and Jahuda (JQR, XV, 708). I am convinced that the current equations of the sibilants in the Semitic languages require recasting, and, unless this is done, biblical exegesis will not make much headway. S In his Kitdb al-Mustalhik (Opuscules, ed. Derenbourg, pp. 234 if.) Ibn Ganah defends the geminate nature of ip t 093 ~, etc., against the , " pp~r pt and giving them all also lpfndetractors of HIayyuj, adding to these " O al- Uztl (ed. Neubauer, cols. the meaning "crush, oppress." However, in his Kitab prIV 742 f.) he suggests also another meaning, "walking to the right and to the left," akin to

L =e, tt=l , WT*= Barth has proved (loc. cit.) that such discrepancies from

the Arabic

~p

u?

to Ibn Ezra and Kimlhi, and it still prevails in our X, * This ambiguity was common commentaries and lexica. 10As in most modern lexica. 11Gesenius, Thesaurus, s.v. t. 12 Lane's Arabic-English Lexicon, p. 1576, col. 2: " it [the lightning] extended high, into the midst of the sky, without going to the right and left."

This content downloaded on Mon, 28 Jan 2013 17:52:28 PM All use subject to JSTOR Terms and Conditions

68

THE AMERICAN JOURNAL OF SEMITIC LANGUAGES

thunder, or at any rate to emit some kind of a noise.

Furthermore, 1

;Q1=

"he uttered, or pronounced, speech in the best manner,"' and

= "he neighed (of a stallion) or uttered a cry (of a sparrow)."2 There is also the testimony of a manuscript supposed to be the Jami' of El-Karmanee is "loudness of voice," or the "being that the primary meaning of X A loud in voice."3 Applying this meaning to the passage in question, we at once get an excellent parallelism: "As a roaring lion and a howling bear is a wicked ruler over a poor people." The same meaning fits very well Nah. 2:5, "The chariots rage in the h ; 1p5 :. I~~bi.n. A.lir' . (like thunders) in the broad ways," giving an streets, they reverberate .. unexpected vividness to the picture.4 It is in the nature of the poet who portrays an action in two immediate but independent clauses either to keep up the same tone or else to ascend in the second phrase, so as to complete the thought and round out the action.5 Such, e.g., is the very next picture in the same sentence where the chariots are compared first with torches and then with lightnings. But such an upward graduation is lacking in our current translations, "rage" being more intense than "run to and from," while "reverberate" restores the proper balance. Similarly Isa. 33:4, * ~. , should be rendered "like the buzzing of the locusts t ti' in their buzz midst"6 will it (scil. in the midst of the nations), presaging pp.'5 confusion and distress, in keeping with the preceding sentence. Finally, = "they shall shout in the city"; the following Joel 2:9, =.i it is evident, ".1i precludes the current rendering "run to and from."7 J.'P. , 1.X That this interpretation has not only an etymological, but also a traditional, basis may be seen from the fact that the Targum translates in a like manner in at least two of the four passages;8 that Rashi, the coryphaeus of
1 Lane's Arabic-English SIbid., p. 1577, col. 2.

Lexicon, p. 1576, col. 3.

3 Ibid. It is not impossible that the common meaning of this root, "break," "split," is secondary and derived from the primary, " utter a noise," since a noise is produced in splitting or breaking. This, indeed, is evident in such expressions as

15
.A .

and 54:1 Koran

84:1. ibid.,

SNotice the appropriateness of an echo in broad ways. 5 Cf. Rothstein, Grundziige des hebreischen Rhythmus, pp. 51 ft. 6 Noise in a lesser degree, just as 0011 is used in various degrees: of the raging of the sea, of the roaring of the lion, and of the groaning of the afflicted (cf. lexica, s.v.). 7 Cf. Rothstein, loc. cit. 1 J1 and to Nah.: T1r JZ1U'"jaT N,201VMl : 8 To Prov.: , M'12t the latter case he certainly confused it with P6: "weapon," as may be "IJIM1M.In m seen from his rendering in Isa.: N npl-T bt TbThand Joel: l brzTXrT nevertheless it is significant that he felt obliged to introduce the blPM; ; "'ITT element of voice.

This content downloaded on Mon, 28 Jan 2013 17:52:28 PM All use subject to JSTOR Terms and Conditions

CRITICAL

NOTES

69

Jewish exegetes, does so in all the four passages;' and, finally, that Ibn mentions the possibility of p 2 being something akin to Ganah. ~.. I am inclined to class here also Gen. 15:2, and make it p2"'1, 'correspond to jP , treated above. This word, pointed as a segholate,3 is naturally ~.'. derived from an imaginary Pt , for which there is no analogy in the Semitic languages.4 But already Ibn Ezra felt the necessity of classing it under the geminate verbs,5 without, it is true, finding a satisor rather , "utter, converse," gives factory meaning. Now k , 3 Y, "one speaking fluently," "a rise in Arabic to an expression "6 to the Aram. spokesman, analogous , "spokesman," N"?21 or 2"] event, "representative, leader."7 We might therefore point ptn 8 and render "the spokesman (or leader) of my house." Different from these is the expression ~j?1 which occurs in Isa. 29:8 and Ps. 107:9. Its meaning is so p. evident from the context that all commentators and lexicographers agree in trans(!l1 ~ l) =f lating it by "longing." But its etymology has always been a matter of speculation. However, admitting, as we must, that it is akin to in meaning, it likewise should be derived from desire," "long, ;.16PI= P. . ; .9 Accordingly, and not, as is generally the case, from = "&* ppr' is a participle of the type P61'1land, like op n, represents the Pt'l and is intensive.10 This construction is very appropriate: the longing Pi., t..,
=1 2 , A t 1 To Prov.: mpf 52 p1 ' ; ]b '. Joel: ?3 to Isa.: ~7? 7 n 2 Kit4b al- Uzfd,ed. Neubauer, col. 743: &Ja
So, already, the Septuagint
o vinb roi^roiTovrov oLsdav

in

; to Nah. and

p
. Confrast, J,4 however, Aquila's

U,4'

whose

MaaeK=psc

(Symmachus: noncommittal. 4 The etymon , which is treated further, below, apparently has nothing in common with this word. 5 nb r'n r r i. Cf. his commentary ad loc.: pPI0) 1Int 6 Lane's Lexicon, p. 1579, col. =1 1. 7 Cf. especially b. Sanh. 14a and Ket. 17a: 'j"3IntnCi :'j1 , "leader of his That '212 "lead" and "1'1 "speak" are cognates with a derivative meaning people." has been demonstrated by Gesenius (Thesaurus, p. 313) and others. 8 As pointed out above (n. 3), Aquila's wv points to the pronunciation with roori pathah. Perhaps the same reading underlies the Targum: bo) g 1") 1, followed by Rashi. 9 Probably due to its similarity with =1. 10 Cf. Gesenius-Kautzsch, Grammar28, Eng. transl., ? 72m, pp. 197f. pp.

The other two translators . 6 oSvyyevihs iou oltsov ov, andl'-73q io' Theodotion: 0 vibe isrl ~ie oltisag ov) are
ov, implying

This content downloaded on Mon, 28 Jan 2013 17:52:28 PM All use subject to JSTOR Terms and Conditions

70

THE AMERICAN JOURNAL OF SEMITIC LANGUAGES

of the thirsty being more intense than that of the hungry, an intensive verb is employed. The third root of the series, pdj, underlies Zeph. 2:9, lP .w as generally recognized. But what is the meaning of . ? The .-;i"'.*, Jewish commentators are too vague and ambiguous to deserve credence.' bj-.: The ancient versions render no help.2 Modern exegetes translate "possession, place of possession," without adducing a cognate etymon or satisfying the context.3 Graetz's emendation to upsets the parallelism with t.p4 Mitt . Schwally makes it synonymous with Arab. "place of %z., is which a I would Arab. again only guess. suggest IA', which, growth,"5 as a term of agriculture, is explained by Dozy as follows: "remuer l6gerement la terre autour des racines des plantes avec un 5)', c.-A-d., avec un instrument qui est pioche d'un c6t6 et hache de l'autre, piocher ligerement."6 "cut, dig," and signifies "ditch," an P :; is thus nom. loc. of pt2, T '; parallel to appropriate "pit." Translate, therefore, "ditches of chickpeas and pits of salt."7 JOSEPHREIDER COLLEGE DROPSIE
1 Ibn Ezra, like all modern exegetes, associates it with irI' =p like ~ renders n? n.tir likewise in his Lexicon under i~ "f; but in his commentary ad loc.: *flrjI 1 Kimh.i 'tI ?159 42~1t S
2 AagaortKo of the Septuagint

Il;

Rashi ll)

'

'I

i,

shows that it was no longer understood in those days and that the translators simply tried to help themselves. = " possess," which was invented primarily as an 3The equation p =tt while in itself, is not probable in our case, since explanation of ~n j possible 1 '1, it does not furnish smoother sense and, moreover, fails to satisfy the parallel in the context. 4 Emendationes, ed. Bacher, ad loc. 5 ZA W, X (1890), 188 f. 6 Supplement aux Dictionnaires arabes, II, 594. ?Something like this underlies the Targum (and the Peshitta) rh'bt=l t r and signifying "hidingbeing a parallel to lj)ijMj rnWtn utf -nti, hole, or recess" (Levy, Worterbuch ilber die Targumim, s.v. btljj ). place, lurking

This content downloaded on Mon, 28 Jan 2013 17:52:28 PM All use subject to JSTOR Terms and Conditions