Skip to main content

Full text of "Arabisms in Rabbinic Literature"

See other formats


Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world byJSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 

Read more about Early Journal Content at 
journal-content . 

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 


By a. Cohen, Manchester 

In the Talmud and Midrashim, fourteen words occur 
which purport to be Arabic. They are quoted, usually by 
R. Levi,' for the purpose of explaining words of the Bible 
the meaning of which had become forgotten, or as a basis 
for a homiletic interpretation/ The following is a list of 
them. For reasons to be afterwards explained, they are 
divided into two classes. 

Class I 

s!)3V K-\Kl^i6 ]''-\\[> S'3"lW (I 

n^D^DO spnoD^ pip N'mw C3 

K''3D »^2:b riip smw sin •'my pK"!) (4 

snny srir-aS t'T'i^ NUipa (5 

ah^W Kpirb flip Kmjja (6 

>?nTn5? xmTinS p-iip n''3im (7 

x-Tia xpir^ j'niiv Nunj^a (8 

sDtDiB' «-i5?E'S jnip sms?! (9 

s''i3B' Knbi33-in^ fnii^f x'-mpa (lo 

' To R. Levi are ascribed all but Nos. 12 and 13. No. 7 is also quoted 
as an 'Arabism' by R. Eleazar b. Simeon. Akiba quotes No. 8 as used in 
D*n *5*13, and No. 10 is quoted by Resh Laskish as a word found in the 
language of ^'lirj Jp. 

' To the former class belong Nos. i, 2, 6, 9, 10, 12, 13, 14; to the latter 
Nos. 3, 4, 5, 7, 8, II. _ 



Class II 

Nin 'aij? ]wb iw (ii 
^3^' bipE' -IDS [K''3-iyi K-iJn] Ky'D C12 
13 nosD riK n» idk 13SV (13 
Nin uiy tit^^ V3p (14 
These words raise a perplexing problem. In very 
few cases can they be referred with confidence to the 
Arabic; in some cases a resemblance can be detected; but 
in the majority of instances, the words are Aramaic or 
altogether unknown. 

Two theories have been advanced to explain these 
facts. Firstly, it is suggested that xuiy does not mean 
Arabia, but is to be identified with Arrabeh, a Galilean 
town near Sepphoris." Against this it might be urged that 
no reason can be assigned why an insignificant town should 
be specially mentioned for its dialect, and not the country 
or district in which it is situated.' Further, how, on this 
theory, are the genuine Arabic words to be accounted for ? 
The second theory is that the language referred to is a 
dialect of Arabic which contained a large admixture of 
Aramaic' If this were correct, we see from the instances 
preserved for us in rabbinical literature that the predomin- 
ance of Aramaic in this dialect over Arabic was truly re- 

° Adolf Brtill, Fremdsprachliche Redensarten (I,eipzig 1869), p. 29. 

* Thus the Galilean dialect is mentioned in Gen. r., ch. z6, § 7. Bacher 
declares, "Wir horen nicht dass er [R. Levi] ausserhalb des heiligen Eandes 
sich aufgehalten hatte," Agoda der pal. AmorSer, II, 303. It is nowhere 
expressly stated that he traveled to Arabia. But his knowledge of Arabian 
customs (see Earn, r., Proem. 23, ed. Buber, p. 20; Eccles. r. on 12, 8, and 
Tanbuma, nStf'1, § 11) leads one to suppose that he had journeyed to that 

' So Briill, ;. c, p. 40; Kohut, Aruch, s. v. ^D (p. 48); Bacher, I. c. 


Would it not, therefore, be more probable to suppose 
that there was an Aramaic-speaking colony settled in 
Arabia? Modern discoveries have shown us how widely 
spread was the Aramaic language in all parts of the Eastern 
World. Endorsements on Assyrian and Babylonian con- 
tracts, papyri from Egypt, and Aramaic inscriptions from 
Arabia testify to this in ample manner/ The Assuan papyri 
disclose the existence of a Jewish colony in Egypt using 
the Aramaic language. There is consequently nothing a 
priori improbable in the supposition that there was a Jew- 
ish colony, settled in Arabia, which preserved the Aramaic 
tongue of the mother-country. Such a dialect would nat- 
urally continue to use many words which had become 
obsolete elsewhere, since the rate of change in a language 
is much slower in the provincial districts.' It would also 
incorporate words from the vernacular of the country. 
These two facts, added to the usual differences of pronunci- 
ation to be found in dialects, will account for practically 
all the 'Arabisms' in the rabbinic literature. 

One important indication that the 'Arabisms' are really 
traces of an Aramaic dialect spoken in Arabia has hitherto 
escaped notice. There is a marked difference in the phrase- 
ology which contains an 'Arabism' and that which contains 
a reference to another language. In Class I we have uni- 
formly flip NUnya "In Arabia they call, etc.'" This 
need not necessarily mean that the word which follows is 
Arabic. When, for instance, the Talmud declares V^IP ?333 
xm apvb (Sukkah 5&, Hagigah 13&), we do not infer 

* See I,id;:barski, Hphemeris fur semitische Bpigraphik, 11, 200 ff., and 
Cooke, North Semitic Inscriptions, pp. 195 ff. 

^ Comp. e. g. the French spoken by Canadian colonists with that of 

* The only exception is No. 4, which will be discussed below. 


that N''3"i is a genuine Babylonian word. The statement 
merely means that, whereas UpV was used in Palestinian 
Aramaic, KUl was usual in Babylonian Aramaic. In 
similar manner N^31'' NlD''Nb pip N''3"i3;3 may only mean 
that Nb^V was the word employed for 'lamb' by the 
Aramaic-speaking Jews of Arabia." 

Further, despite the fact that the rabbinic literature 
contains over twenty references to the Greek language, we 
do not once find the formula piP P'^ "In Greece, they 
call, etc." Instead we meet with such expressions as : 

(Gen. r., ch. 8i, § 5) xin nw ]wb {l^N 
(Gen. r., ch. 40, § 4) xin ''iv \wb dsn 
(Shabbat 63^) Dab 3^3^ tmp ijv jityb 
(Tanhuma, IV) bltt^ N3Dp ■^aoibii t^cb 
(Abodah zarah 246) T31 XIBob ''Xma lip 
(Abodah zarah 24^) KjriK'i mji^-'SDlDllp 

In all these instances, the phraseology indicates beyond 
doubt that a Greek or Persian word is intended. These 
are to be compared with some of the Arabisms contained 
in Class II, but are quite distinct from those enumerated 
in Class I. 

Hence, with few exceptions, the 'Arabisms' in the 
rabbinic literature presuppose the settlement of a colony 
of Jews" in Arabia, where Aramaic continued to hold its 
own against the vernacular. That the migration took place 

" Contrast the two statements jnip Vn 'pnssS tJIsSntfa Na'pj? 'l 10» 
ntO'B'p nS?oS (Rosh ha-shanah 260) and flS D'ttB' 'Enas Bta Xa'pj? '"I Ititt 
D*ntt' *p*TDKa (Sanhedrin 46). In the former case a Hebrew word used by 
African Jews is clearly intended, in the latter a native word. 

'" That they were Jews is evidenced by the presence of Hebrew words 
in Aramaic form. 


at an early date, long before the commencement of the 
current era, may be inferred from the presence of words 
which had passed out of use in Palestinian and Babylonian 
Aramaic. Especially interesting is the apparent survival in 
Aramaic form of Hebrew words which had become obsolete, 
N^nv, n^D^^DD, K^-'II?, and KTiB. 

The following is a detailed discussion of the words: 

Class I 

sSav (.1) 

(p. Berakot 9, i end) i62V KlDixi? imp K'anw '''6 m n»N 
"R. Levi said. In Arabia they call a lamb tobla." Quoted 
to explain ^ivn '\)efOZ nMI (Josh. 6, 5). Comp. VOtyo "'XD 
''n^b^t^'^ to-ipj? 'i idk x«3m ? wn siann m^'b ab^v "sm 
N^3V HMlb imp vn smy^ "How do we know that the 
word tobla is an expression for 'lamb'? For it has been 
taught in a Baraita, R. Akiba says, when I went to Arabia, 
they were calling a 'lamb' tobla," (Rosh ha-shanah 26a). 
Kohut refers to uabilat{un) which is sometimes used in the 
sense of 'lambs,' but more commonly of 'camels.' But the 
word is obviously the Hebrew Sav. We here meet it 
among an Aramaic-speaking colony in Arabia, just as it has 
been discovered on an inscription belonging probably to 
Carthage, which has been unearthed at Marseilles (see 
Cooke, North Semitic Inscriptions, p. 112). The Targum 
only uses the word in the meaning of 'Jubilee' (see Levy, 
Chald. Worterb., I, 325). In Syriac, the meaning 'lamb' is 
not found, but in the Syrohexaplar b^V is used for D''^3Vn 
in Josh. 6, 4; but that merely represents Greek f"|3»7?- (i. e. 
Hebrew hlV in transliteration) which Origen took over 
from Aquila. 


nS T'K ,iSSt d'^^js i^e'iD me'a hmi wnnje* ,d'^S33 DDinSi 
NDH^ S^E>3^ t''"'''P ^«''3■l5;3 " 'And their flesh as dung' 
(Zeph. I. 17) : because they were slain, and their 
flesh was cast upon the ground like dung. R. Levi 
said, In Arabia they call 'flesh' lahma" (Exod. r., 
ch. 42, § 4). The explanation here given of Din^ agrees 

with the LXX am rac aapKag avTuv. QJlh is not found in 

Aramaic in the sense of 'flesh.' Here it is to be explained 
as a loan-word from Arabic, or rather a loan-signification. 
It is not strange that a colony speaking the Aramaic lan- 
guage in Arabia should have adopted that signification, 
especially as lahm and nvh frequently denote 'food' gen- 
erally. Comp. net* Dn^ Lev. 3, 11-16, where the flesh of 
the burnt-oflfering is intended. 

""6 -i"t< .ppiD 1D3 nbo D"''i ,(Lam. I, 15) 'n n'as b:> nbo 
n^D^DO xpnoaS pnip x'ans'^ "There are some who ex- 
plain silla to mean 'comb.' R. Levi said. In Arabia 
they call a 'comb' mesalsela" (Midrash, ad loc). In 
Arabic salla means 'to extract,' and there is no trace 
of the meaning 'to comb.' We find musalsal(un) 
'a piece of cloth figured with stripes or lines' and 
mutasalsil(un) 'a piece of cloth woven badly' (Lane, 
p. 1398), which are probably borrowed from Aramaic 
(Frankel, Premdworter, p. ^6). Comp. Gittin 59a where 
nl^D^D is used for a linen garment. The explanation of 
n^D^DD is to be found in the Hebrew h>tho 'to curl the 
hair' Nazir i, i (comp. 3a). The following passage from 
Rosh ha-shanah 26a is instructive : nboho 'KD pm ijn'' 1in si) 
Nnaa xinni) mos mim -im ••an snoxi) niyoe-' in ndv .noDi-im 


"The Rabbis did not know the meaning of nhobo in Prov. 
4, 8. One day they heard the maid in Rabbi's house saying 
to someone who was combing (curling) his hair, 'How long 
wilt thou ^D^DD thy hair ?' " The fact that the Rabbis were 
ignorant of its meaning shows that the word was obsolete 
and only used in the country villages, from one of which 
the maid possibly came. It is consequently interesting to 
find the word survive among the Jews of Arabia. 

noK lib •'3-1 ,{^nipn nna Diaioc n»t<'2J b^ in-ax n-'n^ .niE' >3k 
(Lev. r., ch. i, § 3) ki3D K''33b |nip 'aiw Nin my p^'^ 
'"The father of Soco' (I Chron. 4, 18); (so called) 
because he was the father of prophets who look (into the 
future) through the aid of the Holy Spirit. R. Levi said, 
It is an Arabic expression; in Arabia they call a 'prophet' 
sakia." There is no such root in Arabic. In Syriac we 
have sakkl 'to look out for, await' ; and in Jewish Aramaic 
the root is very common in the sense of 'to look.' The use 
of the word in Aramaic for 'prophet' is paralleled by bxiDC 
nsWD (I Chron. 26, 28) and nxiao nJ {ibid., 29, 29), see 
Levy, Chald. Worterb., II, 162. Comp. the use of nnn, 
nxn, and naix in the Bible. The presence of the words 
Nin »any pcb, as already indicated, presents a grave dif- 
ficulty to our theory. It should, however, be noticed that 
the formula [nip 'sa Nlrt 'a |1E'5' occurs nowhere else in 
the Midrashim. Unfortunately there are no parallel pass- 
ages to aid us ; but possibly the original version read simply 
imp KunyD ids 'ib •'an. 

Nnnv (6) 

03^ innj mn n»3 i»k tnjv 't ,ib hdik no Tiiyx no 

(Midrash to Lam. 2, 13) KDHJ? ^nnb fnilS N^aiw ^lb 1"K 


"R. Jonathan explains ITys as "How many spoils have I 
given thee." R. Levi said, In Arabia they call 'spoil' 'adlta." 
The lexicographers refer the word to 'addmt(un) which, 
however, means 'enmity,' not 'spoil.' Nearer to hand are 
the biblical 15? 'prey,' and the Targumic Nfinj; Prov. i6, 19 
and ny Isai. 33, 23, etc. 

N^''1V KpU'b pllp K»D-IW ■')b T'N ,tin''p''J' .Dn^iSy )S«S3 m^B" (Gen. 
r., ch. 36, § I, Lev. r., ch. 5, § i) "'They send forth their 
little ones like a flock' (Job 21, 11) : 'ami means 'young.' R. 
Levi said, In Arabia they call a 'child' 'autla." Although one 
naturally thinks of 'ital(un) and 'aw{l(un), it is doubtful 
whether we have here a true Arabism. For the Arabic 
word means 'the persons fed,' and can refer equally well 
to a wife or a slave; it signifies 'family' rather than 'chil- 
dren' (see Lane). The root is, of course, different from 
biy 'to suckle' which is found in Hebrew and Syriac, for 
that corresponds to the Arabic gala. The word KT1V I take 
to be a survival in Aramaic of the Hebrew b'lV which 
occurs in Job 19, 18 (where the Targum has NV^') and 
21, II. 

KniTIJ? (7) 

'b)^-<b V3K hv nmn n'ntj> doIjd p!? htm ids ^)b 'in ,'n 1^ -inyi 
Noijjn ptj' ,pnD» -inin nn n^jsaD nmn nt n'ni int b^ kib'^ 
NnTTO Nnrnn^ piip (Gen. r., ch. 63, § 5) "'And 
the Lord was intreated ('tr) of him' (Gen. 25, 
21). R. Levi said. It is like the case of a prince 
who, with his father, was digging (htr) for a 
pound of gold. One dug from within and the other 
from without; for in Arabia they call 'digging' 'atirta'' 
The following are the variants: liVDC '1 12 itJJ^X 


Nn-iTil? KmTinb pnilV S'-aiW "lOS (Lev. r., ch. 30 § 3) ; in the 
name of R. Levi niTiv nmnb pniiv K''3ij?3 (Ruth r., ch. 5 
on 2, 14) ; and in the name of R. Eleazar b. Simeon 
nmnv nmnn^ pnw N'aiya (p. Sanhedrin 10, 2). There 
is no Arabic word resembling this with the meaning 'to 
dig.' The explanation I offer is this: we have here merely 
a dialectal difference of pronunciation. The Aramaic- 
speaking Jews of Arabia pronounced the n like an ])• 
We may compare Kl^j; nynh pnilXT snnx JT-K (Lam. r. 
on 2, i), s''is s-iinb pniix ab^hi^ (Gen. r., ch. 26, § 7), and 
Kran 'adder' with war. 

NTID (8) 

Npirb rmiv x^a-iw 'li' '1 "iox ,D''tD3E'n i^k .D'snaa niki 
KTisi (Gen. r., ch. 87, § i) "'And I beheld among the 
simple ones' (Prov. 7, 7) ; these are the tribes [the children 
of Israel]. R. Levi said, In Arabia they call a 'child' patia." 
Tia -wib piip K'anw pc iVJ 'fia mo ,-121 bsb poK' Tie 
(Exod. r., ch. 3, § i) "'The simple believeth every word' 
(Prov. 14, 15); how know we that peti means a 'child'? 
Since in Arabia they call a 'child' peti." Elsewhere it is 
stated in the name of R. Akiba, NTiD Npirb D'n •'DiDa piip ptr 
(Sanhedrin 110&)." The word is usually identified 
with fata(n) 'a youth' or fatin(un) 'youthful.' It is, how- 
ever, possibk that the Hebrew TiB translated 'simple' also 
means 'youthful.' In the first passage quoted from the 
Book of Proverbs the parallel word is D'Jn, and the LXX 

^ By a'n '313 is meant Asia Minor, the dialect of which is mentioned 
several times. In two cases genuine Arabic words are quoted as being 
used there, probably loan-words, not? nni D'fl '3133 t!" DSltD jax "There 
is a precious stone in Asia Minor, called mi" (Megillah 12a), which is clearly 
durrafCun) 'pearl'; and mV p'pi B»n '3133 t!" ins e\11! "There is a bird 
in Asia Minor, called p'p " (Shabbat 21a), which is probably to be identified 
with ^aK««) 'crow, raven' (Dozy, SuppUment, II, 420). 


render vtimov . The root-meaning is 'to be spacious,' and 
as applied to the intellect 'open-minded, simple.' The word 
was then referred, as in Arabic, to the 'young' who are 
simple in mind; and, as pointed out above, there are pass- 
ages in the Old Testament where the rendering 'youthful' 
would be quite suitable for 'na. Therefore it is not improb- 
able that in our word KTiS is to be traced the Hebrew 'ns." 
On the other hand, I notice that Payne Smith (col. 3333) 
gives nn^'' ma 'multiplicatae sunt proles ejus' ; and it is 
just possible that NTia was used in Aramaic for 'increase, 

NBtD''E' "iVE'^ " 'Who hath cleft a channel for each hair [sic; 
Eng. vers, the waterflood]?' (Job 38, 25). R. Joshua of 
Siknin said in the name of R. Levi, In Arabia they call 
'hair' sitfa (Tanhuma, Tazri'a, § 8). 
(Lev; r., ch. 15, § 3) KBD'ty inv^h ]^m'\)n p-inN n'^a n^aia ^2-\ ibk 
"Rabbi Berechiah said, There are places where people call 
'hair" siffd." 

This explanation of f\ti^ is based on a passage in the 
Talmud (Baba batra i6a), where mvon |D 3VN nx mn' ]m 
is taken to mean "And God answered Job by the 
example of the hair." «D'3i KD''3 Sdi 01X2 TiKlD l'D''3 nain 
'1J1 now 'JBa KDU nb Tixna "Numerous hairs have I created 
in a man ; but for each hair have I created a separate follicle 
(from which it has growth)." This word XBD"'tJ' 'hair' is 
otherwise unknown. Kohut conjectures a connection with 
Arabic sbt, with transposition of letters. But sabit(un) 

" It is not denied tliat XTID and fata(n) are the same word. This, 
however, does not necessarily imply borrowing, since they occur in kir<3 Vd 


means 'lank,' and its connection with 'hair' is limited to the 
phrase sabitu-s-sa'ri 'a man having lank hair.' More prob- 
ably the word is to be referred to the Hebrew and Aramaic 
root siDC 'to overflow,' it being descriptive of 'flowing' 
locks. We might, perhaps, compare mvc 'storm' and "lyc 
'hair,' lit. that which is tossed about by the wind. It is 
true the Oxford Lexicon distinguishes the roots, but the 
Assyrian sarfu 'hairy skin' and sdru 'wind' (Delitzsch. 
Assyr. Handwb., p. 635) seem to be in favor of a common 
root. The Arabic sa'r(un) 'hair' is to be considered a loan- 

N'laB' (10) 

N''1^S^' Nn^unn^ "Or who hath given understanding to the 
sekul (Job 38, 36) ?' That is, the cock. R. Levi said. In 
Arabia they call the cock sekueid (Lev. r., ch. 25, § 5). 

In p. Berakot 9, i, near end, the reading is |inilS '0n3 
•'1355' N^UJin^ "In Rome ( ?) they call, etc." And else- 
where it is stated, sncj [p Dinnb insl^nTO B"p^ p C'l idn 
'IDS' ^unn^ }nip vn (Rosh ha-shanah 26a). Only in rab- 
binic writings is ''13B' found in the sense of 'cock,' and it 
seems to have been in use in certain provincial dialects. 
Here we find it in an Aramaized form. 

Class II 

nya (n) 

Kin >\> nns sno-iD n-h '-w ps ,xin "anv ntf^ ""h i"« nnovb npno 
'is nV30 noiN " 'Behind {mibha'ad) thy veil' (Cant. 
4, i) : R. Levi said, It is Arabic. If one wishes to 
say 'make room for me,' he says 'nyao for me.' " (Midrash, 
ad loc.) R. Levi wouM render the passage in Canticles "take 


away, remove thy veil." The reference is, of course, to 
Arabic b'd. 

nn" C12D 

»i2V n:n na -13 nn -idx .nan-' 'n ^j; il^B'n ■>«» pan ••yT' iin s^ 
^ipB* 'b noNi KJ1D Njm nin xsji'D sinn nni xj^irs mn in 
'kIjdjx iin l^rr' "The Rabbis did not know the 
meaning of 'cast upon the Lord thy tehab' (Ps. 
55, 23). Said Rabba bar bar Hana, One day I was joui-ney- 
ing together with an Arab, and I was carrying a load. And 
he said to me 'take thy iehSb and cast it upon my camel" 
(Rosh ha-shanah 26b). 

The reference is to tfahb(un). Since the word sy'D 
is used for 'Arab,' it might signify one of the Southern 
Arabs, who said tahb instead of mhb. 
t» p^'iD }in3E' NHB^n -13 JIVCE' '11 '1 13 tWOE' '11 .131 H^T] '1 

ii'S' 5«3^ -ton P n-iraSo^ x''3i5/i Nun xinS pnxi DUinn 
iitj'D san' n-'j'D ppDEj* /^'P xan'- pin 'i^n nn3n!5 noxi 
"R. Hiyya the Elder, R. Simeon b. Rabbi, and R. Simeon 
b. Halafta forgot the meaning of certain words of the 
Targum and went to a merchant from Arabia to learn it 
from him. They heard him say to somebody, 'Place this 
tahba upon me'; whence they learnt that lahba means 
'burden'" (Gen. r., ch. 79, § 7). 

The merchant who is mentioned need not necessarily 
have been an Arab. Possibly he was a Jewish merchant 
from Arabia. Note that in the continuation, it is rri^p J?DK' 


D3 (13) 

iDi!? nxii HMi n p1D3d ns no nnani' loxn ■'3is?i n^i^p ua^ nij?i 
diyc-i DniDj>l DTiai ,''3 ne'VO ns .id "They further heard 
an Arab say to someone 'why dost thou mekasse me ?' He 


meant to say 'why dost thou crush {me'asse) me?' As it is 
written 'And ye shall tread down {'ss) the wicked' " (Mai. 
3, 21) (Midrash Gen., ibid.). 

The word is kassa 'to grind,' maksus(un) 'bruised, 
pounded.' The pronunciation of the 3 and V was somewhat 
similar. Comp. sa^j? n3''3^ pniisn xnnx rr-x (Lam. r. on 

2, I). 

Wp (14) 

Ds? wt^fnb K3 ■'IIP .xin ■'ins? tic^ 'i^ T'x .D'h^s ms j?3p\T 
irpaip nsn iJ^^ni nxn i^ nDS' iT-an " 'Will a man rob 
(kb') God?' (Mai. 3, 8). R. Levi said, It is Arabic. When 
an Arab holds conversation with another and wishes to say 
to him 'Art thou he who defrauds us ?' he says to him, 'Art 
thou our kobea'f" (Tanliuma, Terumah, § 7). Comp. the 
following: [yap rrb nos ,n'DpS xi3J ndk sins sinn!? j;V''K ^"h 
\\>ty iT^ noK KKmo 'n h^^w khk ■T'!? -losp ■'XD j/T' nin s^ .x'j^si 
"Levi [b. Sisi] came to a certain place. A man stepped up 
to him and said to him 'So and so is a kab'an.' He did not 
understand what he said. When he went and asked at the 
school they said, 'It means thief" (Rosh ha-shanah 2a-b). 
The Arabic kb' has quite another signification. I suggest 
that the word intended here is kbh, which was either pro- 
nounced like kb' (comp. No. 7), or was used dialectally for 
it (see Wright, Comparative Grammar, p. 48). The refer- 
ence is perhaps to the common phrase kabahahu-llahu "God 
deprived him of the attainment of his desires." R. Levi 
accordingly explains D'n^s ms Wp^^ as being the reverse 
of this — 'can a man deprive God of anything' in the same 
way as He can deprive man?