Skip to main content

Full text of "The Fragment of the "Sefer Ha-Galuy""

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 by JSTOR. 

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 



The editors of the Jewish Quarterly Review forwarded to me, 
with the author's consent, a proof of an article on the above subject 
by Prof. Margoliouth, asking me to subjoin any counter-arguments 
I may have to bring forward. Such confidence in the editor of the 
fragment of the Sefer Ha-Galuy makes the latter's duty all the more 
stringent, to discard all preconceived notions in favour of the text 
edited by him, and to approach the question in a purely objective 
spirit. I hope that my readers, and even Prof. Margoliouth himself, 
will admit that I have, at least, endeavoured to be strictly impartial. 

The result arrived at by Prof. Margoliouth in his learned and 
ingenious inquiry is astounding. It amounts to this, that the 
fragment is no fragment at all, but a fabrication by some Karaite, 
composed after the year 962, and intended to serve as a lampoon 
directed against Saadiah Gaon, satirically imitating and parodying 
the latter's philological method and style, and inserting some of 
Saadiah's opinions. I call such result astounding, because no scholar, 
and probably up to the present moment not even Prof. Margoliouth 
himself, has ever detected this personation 2 . 

But this can scarcelyserve as an argument against Prof.Margoliouth's 
assertion, living as we do at a time of most surprising discoveries 
of monuments and MSS., buried in the earth. Why should not, 
for once, a discovery be made in a printed book ? It is, therefore, 
purely and simply a question of demonstration. In one respect we 

1 {Editorial Note. — Dr. Harkavy wrote the following reply to Prof. 
Margoliouth on the basis of an unrevised proof of the latter's article. 
Dr. Harkavy was unable to introduce the slight changes in detail which 
the revision of Prof. Margoliouth's proof entailed.] 

2 Prof. Margoliouth, at the end of his article, refers to an eminent 
authority on Judaeo- Arabic literature, namely, to Stein Schneider. But 
he does not notice that the latter expressed his doubts before my edition 
appeared. Afterwards he discussed some personal points occurring in 
the volume of my Studien und Mittheilungen, which deals with Saadiah 
(Berliner's Magazin, 1892, p. 260), but he does no longer speak of falsifica- 
tion. Nor did the late J. Derenbourg, whose special life-study Saadiah 
was, ever conceive the slightest doubt as to the genuineness of the fragment 
edited by me. 



must certainly do justice to Prof. Margoliouth, lie has not made the 
matter easy for himself. He has, industriously and sagaciously, 
collected a whole arsenal of weapons, and has brought forward a 
number of historical and philological objections, tending, in his 
opinion, to cast suspicion upon the fragment of Ha-Galuy. He 
also, honestly and candidly, has produced his whole critical apparatus, 
particularly such points as were taken from Saadiah's own writings. 
But there are a few things which Prof. Margoliouth has omitted to 
do, to the detriment of his inquiry. In my monograph on Saadiah's 
Egron and Galuy I gave, in the introduction to the -former, all 
historical data known to me, and quoted all philological data in my 
notes to the text. The latter, however, were not subjoined to the 
Galuy, because the society Mekitze Nirdamim hurried on the publica- 
tion. But, as will be shown below, the short text of the Galuy can 
be explained from references produced in the Egron. It is much to 
be regretted that Prof. Margoliouth, who so thoroughly discusses the 
second part of my work, entirely disregarded the first. He might 
have saved himself much trouble. Besides, he ignores also, inadver- 
tently, of course, several data produced in my second part. 

Before entering in detail upon the proofs brought forward by 
Prof. Margoliouth, I must preface two observations regarding the 
standard pf judgment and the methods applied by him. As to the 
standard of morality he applies to the Galuy, we find that he compares 
quantities which are altogether incommensurable. He contrasts the 
calm and delicate tone employed in such works of Saadiah's as are of 
purely scientific character, to the violent, irritable, and, frequently, 
indecorous tone met with in the Galuy, without considering how 
different Saadiah's position was in either case. It is in most cases 
easy for us, who write our works, even our polemical writings, whilst 
seated in our comfortable studies in complete tranquillity of mind, to 
preserve calmness and politeness. This Saadiah also understood, and 
acted accordingly in his scientific works. But now consider the 
position of a man who, having risen, through his merits, to the 
pinnacle of social distinction, became involved in a struggle with a 
dishonest but powerful opponent. He succeeds at first in overthrowing 
his enemy, but afterwards succumbs to the latter's unscrupulous 
machinations and those of his party. Insulted, and even personally 
maltreated, Saadiah is obliged to wander about homeless, compelled 
to hide himself to save his life. This not being enough, mud was 
thrown at him in public manifestoes and lampoons; the filthiest 
slanders were levelled at him, and he was threatened with moral 
annihilation. Are we entitled to expect such a man to preserve 
polite and parliamentary speech when replying to such opponents '? 


What should we think of a musical critic, acquainted with Tamberlik's 
and Mazini's melodious song as heard in the theatre, who afterwards 
hearing them howl and screech in an unnatural voice when attacked 
in a forest by robbers and murderers, would come to the conclusion 
that these cannot be the same persons? On such occasions it is 
always as well to think of the Talmudic apophthegm DSfU DIN J'N 
njW TW3 (Baba Bathra, 16 b). 

In another respect also, the standard applied by Prof. Margoliouth 
has not taken the right direction; namely, in regard to the philo- 
logical side of the question. Of course, if we were to take into 
consideration the newest edition of Gesenius' Hebrew Dictionary, 
together with the most recent comparative grammar of Semitic 
languages, bearing in mind at the same time our conceptions of 
style and poetical composition, it would be very easy to show that 
the fragment of the Ha-Galuy, seen in the light of the above-mentioned 
guides, appears to be an abnormal and tasteless production. But the 
question obtains quite a different aspect on considering that we deal 
with a product of the beginning of the tenth century, when Chayyug's 
great discoveries as to the triliteral nature of Hebrew roots, and the 
verba quiescentia and defecfiva were still entirely unknown ; when the 
Arabic-Spanish school of poetry had not yet arisen, when the Paje- 
tanim were still the only masters of Hebrew poetry, and Kalir's 
productions were held to be standard works. That such were the 
conditions of the time can be seen from the first half of my work, 
cf. infra, and they account for the character of the Galuy, which thus 
presents nothing strange or striking. Considered in the light of the 
grammatical and stylistic knowledge of the time, it rather turns out 
to be a tolerable poetical production. More than this. Even if 
Saadiah's name had not been mentioned, an intimate acquaintance 
with Saadiah's grammatical and exegetical views, and with such of 
his works as have already been printed, should be sufficient to point 
to him as the author of the fragment of Galuy. All this will be 
further shown in detail. 

After these preliminary remarks, we proceed to examine Prof. 
Margoliouth's objections to our text, and to see whether they really 
possess the value he ascribes to them. I shall observe the same 
order as he follows in his article. 

I. Prof. Margoliouth asserts that, had the Galuy been translated 
into German or English, it would not have been taken notice of in 
connexion with the Ben Sira controversy ; but since we possess it 
only in the Arabic original, and in a faulty Hebrew translation, 
alleged to be the work of a relation of Firkovich, the difficulty to 
form a judgment is considerably increased for many who are 


interested in Ben Sira. I am very grateful to Prof. Margoliouth, 
that, in spite of my distinct notice (p. 149, and note 2 ibid.) that the 
Hebrew translation is mine, and that, at the time, I had not yet 
before me the one compose.d by a grandson of Firkovich, he yet 
asserts the contrary, probably, for the purpose of attenuating my 
fault. I have committed some errors of translation ; these I have 
afterwards noted myself for the most part, and were at the time also 
noted by others (Prof. Bacher and Dr. Porges in R. E. J.). Not 
a single error of mine has, however, hitherto been discovered, which 
bears any reference to the genuineness of the Arabic original, which 
after all should be of the most importance to Prof. Margoliouth. But 
if no error of that nature occurs in my translation, and the Hebrew 
tongue in which it is composed must be known to all those scholars 
who alone can have a voice and a vote in the Ben Sira and Galuy 
questions, it is difficult to understand why a German or English 
translation of the Galuy would have been of particular use for 
Prof. Margoliouth's assertion. Can the Kreihi and Plethi, can those 
who possess English and German but have no knowledge of Hebrew, 
express an opinion on such a complicated question, which enters so 
deeply into philology and Hebrew etymology ? And what value would 
such opinion have for a man of Prof. Margoliouth's strict philological 
training ? 

2. Prof. Margoliouth looks for, and finds, support in an Arabic 
author, Abulfaradsh Ibn-Alnadim, who wrote about 987; his work 
Fihrist contains also additions leading up to the year 399=1008-9 
(vid. Fliigel's preface, p. xii). This author gives a list of the 
Hebrew canonical writings, and another list of Saadiah's works. The 
former he says that he received from one of the most distinguished 
Jews (Dn^ifNDN ]ti), but he does not give the authority from which 
he derived the second. 

Whosoever knows anything about information from Arabic sources 
as to Hebrew literature should be able to gauge the value that is 
to be attributed to it, even if produced by otherwise trustworthy 
Mahommedan writers. It would be certainly an easy matter for 
Prof. Margoliouth to compile a thick volume about curiosities of that 
kind. Besides, if we wish to utilize the notice in the way of proof, 
we can only do so by accepting Prof. Margoliouth's many conjectures 
at the same time. Both lists of Abul-Faradsh were presumably given 
him by the same Jew. This Jew, who praises Saadiah, was pre- 
sumably either a pupil or a follower of the latter. But a pupil or 
follower of Saadiah's would presumably also have known Ben Sira 
if the master had made use of him. If therefore that pupil had 
known anything about Ben Sira, he would presumably have given 


information about him to the said Arabic author, who would have 
given him a place in his list. 

But apart from the many-storied construction of this hypothesis, 
which deprives it of all force of demonstration in a controversial 
question, all its constituent parts are improbable in themselves. In 
the first place, the two lists can hardly have originated from the 
same source ; for the Arabic author designates in the first list 
" a distinguished Jew" as his authority, but in the second list, which 
follows immediately after, he refers to the opinion of "the Jews'' 
at large (linvN DyW), and not to that of "the same Jew," as is 
the custom with Arabic authors. In the second place, both lists 
scarcely originate from a pupil or learned follower of Saadiah's. As 
to the first list — apart from linguistic blunders, such as, for instance, 
p!D3K (pi. nKplD3K) for p1DB, "IKB3 (pi. DN1ND3) for mt32n, ''DSD 
for D'DSIt?, WO for DWO (probably not status constructus as 
Fliigel conjectures, but an Aramaic form), &c. — the statement that 
the Torah consists of five parts, each of which is divided into two books 
(flSD ^N DD3 i>3 DDpri DND3K riDDi Nil nNTin^K), can hardly 
have been made by a learned Jew. Nor is it at all possible that 
the notice, that Moses was the author of the Mishna, can have 
originated from a pupil of Saadiah's who was acquainted with the 
latter's writings, for Saadiah himself distinctly names Jehudah 
Hanasi as the author of the Mishna. Flugel's conjecture to Fihrist 
(ii."2) that Deuteronomy = min i"WD is meant here, is untenable; 
for, firstly, the Arabic author had already mentioned the five books 
of Moses, and, secondly, it would contradict the description of the book 
as given in Fihrist, where it is stated : " the Jews take from this book 
their jurisprudence, the laws and the sentences ; it is a comprehensive 
work in the Chaldaean and Hebrew language" (TlHvN ronD" 1 i"DC1 

"ON-oyi voids nmb) -raa asm vn OKanniw jwiB^tn npsbx abv)- 

This description can only apply to Mishna and Talmud, as Prof. 
Margoliouth also assumes. 

The second list can, as little as the first, belong to a pupil of 
Saadiah's or to anybody who was intimately acquainted with his 
works, for it contains several absurdities, which, evidently, have 
their origin in ignorance and misunderstanding. Thus we know 
now, that Saadiah wrote, in the first instance, the Arabic translation 
of the Pentateuch, accompanied by a very diffuse commentary, of 
which latter fragments have been preserved in MSS., and quotations 
in Rabbinite and Karaite writings. Subsequently there arose the need 
of a translation only, without a commentary. Saadiah himself names 
the former work in the preface to his translation (ed. Derenbourg, 
p. 4) TOsta mir^N TDBn 3Km, and the latter (ibid.) T'DSJl 3«n3 


min^N fWDa. The latter is described in Fihrist as : T'DDn 3NT13 
mK> tbz NpM nNlin^N. But instead of the former we find in the 
list something which makes no sense : " Explanation (or translation) 
of the third book of the second half of the Torah with commentary" 

(nne>o riNTinta jd n5xi>N ^bt< |D niwri^N nsD^N i^dbh ittnz). 

This could only have been written by an ignorant man, who had no 
knowledge of Saadiah's chief exegetical work ; for, in the first place, 
the Pentateuch is not divided into halves (fpU) but into books 
(1NDDN, 1BD). Secondly, it contradicts Abul-Paradsh's own notice, 
as given above, that each of the five books of the Torah contained 
two books ; consequently, the third book can only be the first half 
of Exodus, which, again, cannot belong to the latter half of the 
Torah. Thirdly, we know that Saadiah's large commentary com- 
prised the whole of the Pentateuch, a fact of which a pupil or 
follower of Saadiah's cannot possibly have been ignorant. Again, 
we find in the list an altogether fabulous book, which in Fihrist 
has the title of " Book of explanation (or translation ) of the sentences 
of David " (UtO DK3nN TDBTl 3Nna). Whatever may be the origin 
of this notice, it is enough to stamp the informant as ignorant in 
Judaicis and not as a pupil of Saadiah. After such examples, we 
cannot be surprised to meet with yet another curiosity in the same 
list, namely, a book of Saadiah's entitled : "Book of parables, divided 
into ten sections " (nsi'NpD ~WV liTl ^NhttN^N 3NT13). Several con- 
jectures have been made as to the origin of this false notice ; at any 
rate, the curiosum belongs to the original of the Fihrist, for the MSS. 
offer no various readings. Consequently its author can scarcely have 
been a pupil of Saadiah. One might object, that, after all, it is 
possible that the Jew had given Abul-Faradsh correct information, 
which, however, became corrupt by the latter's fault. This is, of 
course, possible. But even if we grant this, we are not able to make 
any use of the corrupt notices, and, at all events, the Arabic author 
loses all value for demonstrative purposes, such as Prof. Margoliouth 
is inclined to attribute to him. 

3. Nor can I find any grounds for assuming, with Prof. Margoliouth, 
that a pupil or a follower of Saadiah would have inserted Ben Sira 
in his list. The first list only enumerates the canonical writings 
of the Jews, but the Talmud distinctly excludes the book of Ben Sira 
from the Canon (Babli Sanhedrin, 100 b and Jerush., X, 1), and allows 
only citation of beautiful sayings (NTVvJJO v'D) out of it. The same 
is found in Midrash Rabba, Koheleth, sub fin., and times out of number 
we find, in the old Jewish literature, the Canon quoted as the twenty- 
four books (D'HtW njaiN), to the exclusion of the Apocrypha. 
Saadiah himself, in our fragment of the Galuy, points out that Ben 


Sira, and the other books quoted by him, were secular books, and 
then we are to expect a pupil of his to count the Ben Sira in his list 
of canonical books ! The only non-canonical book mentioned in the 
Fihrist, the Mishna, is only quoted in parenthesi, because it was 
alleged to have belonged to Moses. The Christians, on the other 
hand, always used to embody the Apocrypha in their Bibles, and 
for this reason Ben Sira is in the Fihrist also enumerated among the 
Christian holy writings. 

4. Another proof that Saadiah could not have possessed the 
Ben Sira in the Hebrew original, Prof. Margoliouth believes to 
find in the circumstances that the author of the so-called Chronicles 
of Jerahmtel knew nothing of the Hebrew text and only knew 
the translation. But I do not think this argument to be valid, 
for the author of the said Chronicles lived, either in South 
Italy, according to Dr. Neubauer's conjecture, or, as Dr. Gaster 
recently tried to prove in his learned introduction to the Chronicles 
of Jerahmeel, p. xlvi, in Spain ; at any rate, in Europe. He drew 
his information from Latin or Greek sources 1 . On the other 
hand, as I tried to show (pp. 198-203, a point not noticed by Prof. 
Margoliouth), Saadiah most likely discovered the original of Ben 
Sira, after his dismissal from office, among the hidden treasures 
of the Academy of Sura, together with the Book of the Jubilees, 
and other apocryphal writings, about which discovery contemporary 
information is extant (cf. Kirchheim's Commentary to the Chronicle 
of the Tenth Century, p. 36, and the parallel passages alleged, ibid, 
in the marginal note). The name WVDi (Apocrypha) had therefore, 
at that time, its real meaning in Babylonia. How could it, therefore, 
have been possible, for a European author, to make use of the original 
Ben Sira which was hidden in Babylonia ? He had to be contented 
with European sources. 

5. Prof. Margoliouth creates difficulties for himself where there 
are none, and in spite of my having satisfactorily explained every- 
thing. Namely, Saadiah's words: "I was then in Irak," refer to 
his first visit to Babylonia, at the beginning of the twenties of 
the tenth century, when he arrived there for the purpose of, con- 
jointly with the Exilarch, David, and the chief of the Academy of 
Pumbeditha, Cohen-Zedek, bringing to a conclusion the struggle 
with the Palestinian Pretender, Ben Meir (vide my Studies, &c, 1. c, 
pp. 212-224). Of course, Saadiah returned home after the strife had 

1 I wish to notice, en passant, that the original form vnn (Onias) is not, 
as it is believed, taken from the Greek, but from the Talmud (Megillah, 
10 a, Menachot, 109 a). This name is probably Tlieophar, and identical 
with pnv, only in the sense of a prayer: " God be gracious ! " 


been concluded and Ben Meir defeated. There is therefore an 
interval of thirteen years between Saadiah's first visit (921-2) and the 
time that he composed the Galuy (934-5), as I have fully shown 
(ibid., pp. 145, 229). Prof. Margoliouth takes no notice whatever of all 
this, he puts irrelevant questions, gives useless answers, and invents 
unsuitable chronological data. He also tries to prepare artificial 
difficulties regarding the Galuy fragment, where everything is in 
perfect order, whilst, at the same time, he commits several small 
errors and inaccuracies, which are here of particular significance. 
Thus the date of Ben Meir's letter is not, as Prof. Margoliouth states, 
923, nor even 924, but 1233 of the Seleucidean era = 921-2. There 
is just as little contradiction between the expressions " in Irak," and 
"in Bagdad," as there would be between "in England," and "in 
London." The former is a more general, the latter a more special 
expression. The Arabic Jews always use the term P&OJJ7N for 
Babylonia, in contradistinction to DNE>?N, Palestine (Syria). The 
Arabic geographers also identify the same. Thus e.g. Yakut, in his 
geographical dictionary (ed. Wustenfeld, III, 631), writes : '•'vN tOfll 
i?3K3 pN PN-1J&N3 TNID^N }N "by Mn blt< p DnjJJ fttWm. 
The expression DK> v3 ^3D (I.e., p. 233, 1. 12) does not mean "an 
unknown man," as Prof. Margoliouth translates it, but "of an 
unknown family," " of low descent," in contrast to the preceding 
(1. 11) innatPm *1333, "honoured through his family," "of honourable 

6. The above-mentioned circumstance that Saadiah had found the 
original Ben Sira, in 934-5, in the library of the Academy of Sura, 
serves also as a reply to Prof. Margoliouth's questions : why Saadiah 
never mentions Ben Sira, either in his commentaries to the Proverbs, 
or to Sefer Yetzira, or in his )"IN3XDK,?K 3KJ13, for these works were 
written before 934. The commentary to Sefer Yetzira dates from the 
year 931 (ed. Lambert, pp. 52-76), the JltONDiOX 3KJ13 was written 
in the year 933 (ed. Laudauer, p. 72), and there are several indica- 
tions to show that the commentary to the Proverbs was written even 
earlier than this. It is also possible that the Gaon, on account of 
the aforementioned se»u'-prohibition in the Talmud, had at first 
scruples against citing from the apocryphal work unnecessarily. Afte r 
he had been violently attacked by his opponents, because of his first 
edition of the Galuy, he permitted himself to make use of Ben Sira 
as a weapon of defence, since the book contains irrefragable proofs, 
that already in ancient times non-canonical books had the external 
attributes of canonical writings, and that, therefore, no reproach 
could be made him that he had given these attributes to his " book 


of the Exiled " (for this meaning of Sefer Ua-Galuy, vid. infra). The 
former alternative appears, however, more likely. 

7. Saadiah was of opinion that the punctuation and accentuation 
belonged to the period of the second temple, probably to the school 
of the ancient Soferim. According to the results of modern historic- 
critical investigations, this was an erroneous view, for we can now 
maintain, with tolerable certainty, that our system of punctuation 
and accentuation did not exist before the second half of the sixth 
Christian century. Saadiah, as a scientific man, and also, because in 
all matters religious he took the Talmudic-Eabhinical Judaism for 
his guide, did not attribute any sacredness or obligatory function to 
the punctuation and accentuation, which is not mentioned in the old 
Rabbinical literature, although certain uncritical writers are of a 
different opinion. Thus, for instance, the author of the Manuel 
du lecteur, edited by the late Derenbourg (the real title being JTQriD 
JN^nn), Moshe Hanakdan, the Karaite Jehuda Hadasi, &c, who 
maintain that the punctuation and accentuation were delivered to 
Moses from Mount Sinai. The same view was only recently defended 
with much acumen and learning by Jacob Bachrach, now deceased, 
in a work of two volumes (b"TW DJ? ITOinti'N, Warsaw, 1897). 
Saadiah's opponents evidently embraced this same view as to the 
obligatory sacredness of the punctuation and accentuation. They 
reproached him bitterly with having dared to provide his productions 
with the holy attributes of the ancient Prophets. Apart, therefore, 
from the erroneous view, shared also by Saadiah, that the punctuation 
and accentuation belonged to antiquity, the Gaon's opponents com- 
mitted another important error. Namely, they confused the notions 
of ancient and holy, an error which Saadiah avoids. We see the latter, 
in his interpretation of Scripture, frequently deviate from the con- 
ception of the accentuation, we also find that, although usually 
following the Targum of Onkelos and the Halachic exegesis of the 
Talmud, he very frequently opens up a way for himself, and, in his 
explanation of the text, deports himself in quite an independent 
manner in the face of the Agadic interpretation. Of course, a sound 
critique must, on this point, unhesitatingly side with the Gaon of 
Sura. It is, therefore, surprising that Prof. Margoliouth commits 
here the same error as Saadiah's opponents, and the inconsistent 
Ibn Ezra. He also confuses the notions of old and sacred. The drift 
of Saadiah's argumentation is that, in spite of the antiquity of the 
points and the accents, they are, nevertheless, not holy, as shown by 
the examples of Ben Sira and other secular writings. Prof. Margoliouth 
protests against this, and maintains that, if old, they must be holy 
and must not be meddled with, but if they are not holy, and if it is 



permitted to deal freely with them, in the way the Gaon evidently 
does, in that case they must be young and of late origin. Prof. 
Margoliouth moves in reference to Saadiah in a vicious circle. In 
doing so, he entirely disregards that it is possible to consider the 
whole system of punctuation and accentuation as extremely old, and 
even as traditional, without, at the same time, believing in the 
traditional transmission of the points and accents of every word. 
Rashi's grandson, Jacob Tarn, was certainly not more broad-minded 
than Ibn Ezra, and yet he writes unhesitatingly that the punctators 
and Massoretic writers have committed errors (D^pin tyO. — H1]TOn, 
ed. Filipowsky, pp. 11, 12). 

The evidences which Prof. Margoliouth thinks he has found in 
the Gaon's commentaries to the Proverbs and to Sefer Yetzira for the 
lateness of the points and accents, are without force. In the former 
(ed. Derenbourg, p. 52) Saadiah accuses the new opponents (the 
Karaites), that, whilst the Rabbinites had fixed the number of chapters 
verses, and words of the Bible, of which they had established the 
correct divisions, and noted how many times each word occurs 
therein, not a single one of the new opponents had been concerned 
in establishing the biblical text, the plene and defective, and the 
grammatical changes of the forms of words. The latter term may 
include the various forms of words in respect to conjugations and 
declensions, and also, in respect to punctuation and accentuation, 
for the Massora deals with both. It appears unintelligible how any 
indication of the lateness of punctuation and accentuation can be 
found in this remark. It is true, Saadiah afterwards deals specially 
with the Halachic traditions. But this proves, that, granted even 
Prof. Margoliouth's assumption that by the word n?N3tWI the vowels 
are meant, the latter do not rank the same as the Halacha, and are 
not obligatory like the latter. But this only confirms the view 
developed in the Galuy as the profane character of the punctuation, 
in which there is nothing sacred. But there is no allusion here to its 

Nor can anything be found in the passages quoted by Prof. 
Margoliouth from Saadiah's commentary to Sefer Yetzira, which 
would in any way contradict the views expressed in the Galuy. 
His objections to the method of reckoning the vowels among the 
letters can only be assented to. It is very doubtful whether the 
words NWN"C pnpi, at the end of the commentary, ought to be 
translated, with Prof. Margoliouth, " a grammar of their sacred books." 
We should rather take it, with the editor Lambert, to mean : com- 
prendre le detail des prSceptes (de Dieu) ; for neither in the JVUK, nor 
anywhere else, is mention made by Saadiah of an ancient Hebrew 

VOL. XII. N n 


grammar. Moreover, if Prof. Margoliouth's translation were correct, 
the term JPNIB^K awia piipT or ^tOB^N fab pHp*l would be 
required. But granted even that Prof. Margoliouth's translation is 
right, yet it would be impossible to conclude from the circumstance 
that Saadiah attributed the establishment of the rules of grammar to 
the same scholars who transmitted the laws and the oral tradition, 
or, in other words, to the oldest Talmudical Rabbis, that according to 
the Gaon the points were rather late than ancient. 

8. As to Eleazar ben Irai, I willingly admit, as I have already 
observed (1. c, pp. 204-5), that the reason why Saadiah attributed 
to him a saying of Ben Sira is still unknown. I only wish to add 
here, that there is a possibility, that in the copy of Ben Sira, which 
was in the hands of the Gaon, the first three chapters, or, at least, 
the sheet containing the third chapter, was missing, or that, by an 
error of the copyist, verses 20-21 had been omitted, and that therefore 
Saadiah quoted these verses in the way they were transmitted by the 
said Eleazar. But whatever the cause may have been, the fact that 
this same passage was quoted in the commentary to Sefer Yetzira in 
the name of Eleazar shows that Saadiah had before him a work of 
Eleazar's in which that passage occurred, and that he had forgotten 
the citation from Ben Sira as mentioned in the Talmud and the 
Midrash. Such a lapse of memory on the part of the Gaon in the 
year 931 (commentary to Sefer Yetzira) may very well have recurred 
in 934 (Ha-Galuy). Under no consideration can this accident con- 
stitute a reason for suspecting the genuineness of the Galuy fragment, 
in which we also become acquainted with two verses from Eleazar's 
own production. These latter are all the more interesting, because 
they belong to those literary productions that served Saadiah himself 
as a pattern of elevated style. I see from Prof. Margoliouth's Essay 
that the verse, or verses, quoted in the name of Eleazar, occur also 
in Dr. Schechter's edition of Ben Sira. (That edition is, unfor- 
tunately, not accessible to me, and I must, therefore, refrain in these 
observations from discussing the "Ben Sira" itself.) But I do not 
know in how far the verse in Dr. Schechter's edition is in accord 
with that quoted in Ha-Galuy, and in the commentary to Sefer 
Yetzira, in which the text is in complete agreement (except for the 
unimportant difference that in the latter ?2 occurs, instead of the 
second PN in the former). Prof. Margoliouth completely ignores my 
conjecture (L c, p. 204), that the Eleazar mentioned in Yerushalmi, 
Chagiga (66 d, 73 c), Bereshith Rabba (c. 8), and Tanchuma, Miketz 
(ch. 10), who produced citations from Ben Sira, was, perhaps, identical 
with Eleazar Ben Irai. I will add here two more sentences of 
Eleazar's, which seem to have been taken from Ben Sira : be 1J1DD 

vhn bt) Wttytte> OIK (Babli Pesachim, 119 b, cf. Ben Sira, ed. Bensew, 

xli, 27, t6n pe^po xet*Di torn) ; and mvi> ni>an din ctP* ohy^ 

(Babli Sanhedrin, 44 b, cf. Ben Sira, ibid., XVIII, II : yOQT\ ^ ly 

k& tmpy). 

9. Prof. Margoliouth takes things very easy in respect to the book 
of the Maccabees, which is quoted in Ha-Galuy. He simply ignores 
everything set forth by me (1. c, pp. 208-9). He disregards the 
mention of Megillat Beth Hashmonai in the Halachot Gedalot, to which 
attention was already drawn by Rappoport, and which passage in the 
latest critical edition of that work (Warsaw, 1874, p. 282) was carefully 
noted by me. He only consults the notoriously late recension of the 
Vatican manuscript, into which an obvious mistake has crept, which 
was already pointed out by the editor (p. 615, note 9), and observes in 
a dictatorial manner: "but this is an error." Prof. Margoliouth 
should be a little more cautious in matters of Jewish literature when 
dealing with Rappoport. 

Again, Prof. Margoliouth makes no mention whatever of the 
evidence of Nissim of Ka'irouan in TlWyen IQD (written about 1030- 
50), which was also referred to by Rappoport and by myself. In the 
Hebrew translation of that work, hitherto known, the Book of the 
Maccabees is called , tOBB>n n?JO. But according to the Arabic 
original, discovered by me, the title was "'tween "03 n?JO (Stein- 
schneider-Festschrift, Hebrew part, p. 1 9) ; almost the same as in Halachot 
Gedalot. It is known that, in the latter work, the authorship of the 
book is ascribed to the elders of the schools of Shammai and Hillel. 
Let us now see how Nissim expresses himself about it: tO OJN3 

NfUD wiam Nrva uto npri »b yeit< nct6i> ni nnn -oi }o rrbix 
naein no tkdi »t«DB>n "oa rbitt "mot* nbm bnv jno in xo •>id 
piKoiw jo pht6to tot^N "6y ni nbo NiBD pe>yi nya-it6tt 
tone KTiiio "j^i jns xnao linani Tterc^wi ("I shall not neglect 

to tell in this book the events that happened to the nation, when 
they were in distress and were rescued, with the exception of such 
as have already been described, as for instance the Megillat Esther and 
the Megillat Bni Hashmonai, and all that is found in the twenty-four 
books about the oppressions and misfortunes that have befallen the 
fathers and the patriarchs and from which they were delivered ; for 
all this has been described already and is found everywhere"). 
Consequently, this testimony of Nissim proves that as early as the 
beginning of the eleventh century, the Book of the Maccabees was 
already universally known in eastern countries as the biblical books. 
This fact is confirmed also by old oriental MSS. of our book and of 
prayer-books in which the latter is embodied in connexion with the 

N U a 


feast of Hanucah. I noted such MSS. from Yemen from the beginning 
of the fourteenth century (1. c, p. 209) ; afterwards, I had the oppor- 
tunity of seeing MSS. of that book which date, at the latest, from 
the twelfth century, but belong more likely to the eleventh. Such 
data have, after all, quite a different value from that of an occasional 
phrase of Jellinek's that the book was a later liturgical work, a phrase 
quoted by Prof. Margoliouth with evident complacency. From the 
present standpoint of Jewish science we can rather maintain that 
the Book of the Maccabees was just as little composed for liturgical 
purposes (and in the Middle Ages forsooth !) in order to be read on 
Hanucah as the Pentateuch was written for the purpose of being 
recited at prayers, or the Prophetical books for the sake of the 
Haphtarot, or the Book of Esther for the Purim liturgy. In the 
first place, there is no trace in the book itself of any liturgical use, 
else it would have concluded with some form of prayer for the 
present and the future times (something like "as the Lord has 
helped us miraculously at the time of the Maccabees, so may he," 
&c.) ; secondly, in that case, the liturgical use of the book would 
have been much more extended. But we find that it is not in use 
at all in western countries, and the writers on decisions and ritual 
do not know it ; only Isaiah di Trani, the older, who lived in the 
thirteenth century, quotes it, and even in the east, it was only very 
sparingly made use of in the liturgy. The inference is obvious, that 
it was not written for liturgical purposes but as an historical work. 
Thirdly, up to the present time, we know of no historical book 
written in the Middle Ages in pure Aramaic (Seder Olam Suta, Seder 
Tanaim Ve-Amoraim, and the Epistle of Sherira Gaon were written 
in the mixed Talmudic-Aramaic dialect). For this reason also we must 
assume that the Book of the Maccabees belongs to antiquity. The 
positive assertion of the Halachot Gedalot, that the work belonged 
to the schools of Shammai and Hillel, is therefore by no means 
objectionable ; it is much more objectionable to declare that a work, 
which was already considered as ancient at about the middle of the 
eighth century, was a late compilation of the Middle Ages. Not a 
single alleged indication of a late authorship, which Prof. Margoliouth 
tries to establish by the aid of philology, can be decisive of anything 
in regard to our question, for it is impossible for us, at the present 
time, to maintain categorically, that such and such a word, which 
is at the present day known to us only from kindred Semitic dialects, 
could not have been in colloquial use in Palestine in the first 
Christian century. Many words in the Mishnah and Gemare can 
only be explained by us with the aid of Arabic, but nobody will 
therefore assert, that this points to Mahommedan influence. In 



addition, to this, there is the circumstance that as yet no restoration 
of the text of the Book of the Maccabees acccording to the oldest 
and best MSS. has been undertaken, so that it is possible, that some 
words have found their way into it from the hands of later copyists. 
But if the work belongs undoubtedly to antiquity and not to the 
Middle Ages, it would be historically unjust to demand of Saadiah, 
that he ought to have recognized it to be non-Maccabean, according 
to the light of the critical aids at our disposal. He produces the 
grounds that induced him to attribute to Abraham no more than 
the ideas contained in the Sefer Yetzira, but not the text as it stands ; 
these grounds could be discovered even at his time, but not those 
which prove that the book in question was not Maccabean. And if 
he himself did not believe in a Maccabean authorship of the work, 
he might, for all that, have made use of it as an argument against his 
enemies, believing as they did in the latter. It was, at any rate, good 
enough for them, their reproaches and accusations against the Gaon 
were thus proved to be unfounded. 

10. We have now arrived at that part which should form the main 
point in Prof. Margoliouth's criticism, namely, the philological part. 
I have already alluded to it, but it deserves to be more specially dealt 
with. I called the philological side of the question the most important 
one, because if treated scientifically, and with regard to a Babylonian 
writer of the tenth century in the position of Saadiah, it cannot 
possibly rest on purely personal ideas and considerations of that 
which is fit and unfit, proper and improper, &c. Nor can a 
philological critique base itself upon hypothetical, mere arbitrary 
chronological, or moral combinations, &c, by the aid of which 
a literary controversy can scarcely be finally decided. On the other 
hand, it is easy for so thorough a Semitical linguist as Prof. 
Margoliouth categorically to prove with the aid of documents, from 
a philological standpoint, that which is possible and that which is 
impossible. For the literary monuments of each epoch and of each 
writer are, on the whole, known and recognized by the students. 
We might therefore expect that the chief attack against the Galuy 
fragment would be made from that quarter, should we really have 
a supposititious document before us. For, not even the most subtle 
falsifier of texts has as yet succeeded in imitating his counterfeit 
productions so artificially, that the critique could not detect in it 
some treacherous weak point. Unfortunately, Prof. Margoliouth has 
refrained from making use in his criticism of those decisive points 
which are offered us in the history of the Hebrew language, of the 
development of the Hebrew poetical style, of the observation of 
the mannerisms of the neo-Hebrew writers in general, and of those 


of the tenth century in particular. More than that, Prof. Margoliouth 
forgot, most peculiarly, to take into consideration the philological 
position and the poetical style of Saadiah himself, notwithstanding 
the fact, that these are the most momentous ones in regard to our 
question, and should occupy the principal place. Instead of this, 
he again enters on excursions into remote and vague subjects, and 
is contented with referring to conjectures, and even to incorrect 

I shall therefore be permitted to make some introductory remarks 
on these points : — 

(a) It is known that Abu Zacaria Jahia Chajjug (at the end of the 
tenth and the beginning of the eleventh century) was the first to 
make the important discoveries of the triliterate nature, and of 
quiescence and defectiveness, of Hebrew roots. Until that time, bi- 
literal, and even monoliteral roots were universally assumed, and there 
existed a rule, that radical letters could never disappear in the 
various grammatical permutations of the roots. Consequently, the 
absence of one or another letter in the words of the Bible proves 
that such letter could not have belonged to the root, but was merely 
an additional (servile) one. They therefore considered the verba 
quiescentia and defectiva as biliteral roots. This is not only illustrated 
by numerous instances taken from the Pajetanim, proofs of which 
can be found in the appendices to Zunz's Synagogale Poesie, but 
Saadiah distinctly enunciates it (in the first part of my work, p. 57, 
ver. 25 of the Egron). It was also recognized by his opponent 
Mubashir (ibid., pp. 71-73), by Ben Asher and Menachem ben Saruk 
(m3nD, passim), and Dunash does not object to it. It is easy to 
verify from the above-mentioned lists of Zunz, that the Gaon acted 
upon that principle in his poetry, which, to him, was identical with 
Pijut (vid. infra), e.g. IN (from riNJI ; Win, Num. xxxiv. 7), *U (from 
*JM, thus also Kalir), ?H (for n?n, like Kalir), DJ (for flBy, like Kalir), 
BT/ (for ntPy, like Kalir), IX (for iW, like Jose ben Jose and Kalir), 
flV (for JW, nw, like Kalir), BH (for VTf). These examples Zunz 
took from the Gaon's iTTOy, n«D a^in, and JinntN, which are 
printed in CJ10TP CM1NJ n» >KW> J*3ip (Berlin, 1856). We could 
add now several other instances from Saadiah's TYUytPin, printed by 
the late Kohut; e.g. JH (for iTJfl, like Kalir), pn (for ppn), )"B (for 
HVB, like Kalir), ft (for "VP, like iTTO 1BD and Kalir), Yl (from TV, 
like Kalir and Ben Asher). 

(6) We know now that the Gaon had a deep respect for the 
Pajetanim, Jose ben Jose, Jannai, and particularly for the prolific 
Eleazar Kalir. This appears from the passages collected in the first 
part of my work (pp. 51, 107-110), in which he calls them "the 


ancient poets" (}"6i*6k KIJK^K), and "the excellent poets" (trwta 
pVKQPtt). He quotes examples from their poems in his dictionary, 
written especially for the use of poets (JVUS). We noticed hefore 
that he imitated Jose and Ealir in his liturgical poetry. But in his 
secular rhetorical epistles he also followed Kalir's style (ND'tfl ]W\ 
tW'KDI 'B 31133) ; this is evident from his commentary to the Sefer 
Yetzira (p. 23; where the correct reading seems to be TtiOTCi 'StDflDI, 
from the Talmudical 3Dn, rU'DII). Saadiah's method was the same 
as that of the Pajetanim, who sought to increase and expand the 
Hebrew vocabulary by the creation of new, and often unsuccessful 
forms of nouns and verbs, by the frequent use of hapax legomena, and 
by employing Targumic and Talmudical terms. Besides the great 
number of examples of words of that kind used by the Gaon, which 
have already been given (1. c), and which could be considerably aug- 
mented, we know now, that even in his exegetical works, and in his 
secular Hebrew productions, he did not shrink from the boldest 
interpretations and formations of words. It is greatly to be regretted 
that the little book, Kritik des Dunasch ben Ldbrat iiber einzelne 
StellenausSaadia's . . . Schriften, edited by R. Schroter (Breslau, 1866), 
seems to have been entirely unknown to Prof. Margoliouth. He 
would have found there that Saadiah, e. g. explained "U3 (Gen. xxx. 
11) as TJnn K3 (No. 14), DID (Ps. xvi. 4) as D*TD (No. 18) ; that he 
derived VIVO (Ps. Hi. 9) from fta (No. 20), ^lOt (Gen. xxx. 20) from 
TOT (No. 21) ; that he identified Dnt (Isa. i. 7) with DTJ (No. 23), 
'Din? (Deut. xxiii. 24) with D^n. Dunash quotes, in the same booklet, 
any amount of such monstrous words of Saadiah's, as "nta (from the 
interjection ITn), nnta (from the interjection nn«), CB1K (from D'DK), 
Jfjrtp (from the proper noun JTip), nrfN from TliK, and Tinn from Til . 
Prof. Margoliouth should be consistent, and apply his hypothesis also 
to the Saadiah's liturgical poems, and to the passages just quoted 
from Dunash, and maintain that they are only satires against, and 
parodies of the Gaon's works. Again, to be fully consistent, Prof. 
Margoliouth ought to apply the same assumption to many passages 
in other works of Saadiah's, which he himself recognizes to be 
genuine, e. g. the Commentary to the Proverbs, xxx. 1 (ed. Derenbourg, 
p. 183), where rip" 1 is derived from ?np1, bra from p?n3, JSVWt from 
Tlv'K, and ?N107 from /ID ; also to many other passages in the Gaon's 
works, which are in no way superior to those in Ha-Galuy to which 
Prof. Margoliouth takes exception. We see, therefore, that Prof. 
Margoliouth's method would lead us too far. 

11. Having thus prefaced some general remarks about Saadiah's 
position in reference to language, we can now proceed with the 
consideration of Prof. Margoliouth's philological observations on 


the short text of the Galuy fragment. We shall again see that 
Prof. Margoliouth's criticism of that text is quite untenable. 

Prof. Margoliouth writes : " The first sentence alone contains two 
words unknown to the Dictionary, JIDn and ffitTl ." To this we reply 
that nWO is a hapax legomenon, occurring in Ezek. xxviii. 17, and 
it is quite legitimately used in Saadiah's elevated style. For apart 
from the innumerable examples of the use of hapax legomena and 
rare forms (Arabic "HNU) employed by the Pajetanim and Spanish 
poets, and also in the Gaon's liturgical productions, we find also two 
examples in his brief rhetorical address in his commentary to 
Sefer Yetzira (p. 23), 'iriBDI (Gen. xxi. 16) and 'JVn (Isa. xlix. 22, and 
two more places) ; not less than five instances in the address 
(apparently to the Academy of Sura) in Dunash's Critique, No. 87 : 
np"0 (Exod. xxviii. 17), th<V (ibid., xxviii. 18), T11DNT and 1313' 
(Ezek. xxvii. 16), 0W1N (formed from DD?riN, Exod. xxviii. 19); 
several examples in the preface to jrON (in my work, first part, 
pp. 52-7), e. g. ver. 2 : 7X8*1 (Num. xi. 25) and VIP (Mic. iv. 8 and 
Job iv. 5) ; ver. 3 : npD7 (1 Sam. xix. 20) and iwSJ (Gen. x. 25) ; 
ver. 4 : "IHtn? (Eccles. iv. 13), WB*I (1 Sam. xi. 11 and 2 Sam. xx. 22), 
DDna (Jer. li. 39), and niTTiyn (Deut. xxxii. 35) ; ver. 5 : fljrv 
(Isa. xv. 4) and JTTTI1 (Dan. xi. 4), &c, &c. 

We are also able to give the authority for the other impugned word 
pDn, which occurs in Ha-Galuy as synonymous to 1X1K, in the sense 
of "treasure," "treasured store"; namely, Isa. xxxii. 18, where the 
hapax legomenon fpfTJ is also used synonymous to "l?tt£. In Saadiah's 
translation, (Euvres compl. de Saadia (III, 1896, p. 33), both words are 
rendered by two Arabic synonyms, one of which is, just as in the 
Galuy, formed from the root JT3. It is therefore to be regretted 
that Prof. Margoliouth did not include the third volume of the 
(Euvres compl. among his critical apparatus ; had he done so he would 
have found in the impugned word rather a confirmation of, than an 
objection to, Saadiah's authorship. 

Prof. Margoliouth writes further: "rM,in the next line, is from 
the Targum of Job." In regard to this we wish to observe, that 
granted even that Saadiah considered the word iVJ as Targumic, 
there would yet be nothing surprising in the fact of the Gaon 
making use of it. Saadiah followed the example of biblical poetry, 
in which Saadiah, after Dunash's Critique, declared the words pDn 
(No. 26), mm (No. 27), -ip'3 (No. 40), fO^Q T\Txb (No. 48, cf. the 
Commentary to Proverbs, xxxi. 3, ed. Derenbourg, p. 197) and others to 
have been borrowed from the Aramaic, and in which, after the 
method of his well-known little work, there are over seventy (91) 


words which must be explained from the Targum and the Talmud. 
This was the method he applied to his poetical productions. Thus, 
in his liturgies he constantly makes use of Targumic and Talmudic 
words, and in his vocabulary, fVUK, specially composed for the use 
of poets, he included words from the Targum and the Talmud. Cf. the 
first volume of my work (pp. 69, 71). The impugned word can 
certainly have been used by Saadiah, for Zunz showed already 
(Synagogale Poesie, p. 394) that Saadiah had used the word HPTO, 
probably as a nomen actionis, in a Selicha, and that Kalir also had used 
the word n*? (p. 390). We have already seen that Saadiah held Kalir 
in high esteem and that he was fond of imitating him (cf. infra). 
We would therefore be entitled to assume a priori, that if HiTJ 
belonged to Saadiah's vocabulary, the same may have been the case 
with iTJ. This is now confirmed by the Galuy fragment. But we 
can go even further than this. Zunz and Prof. Margoliouth consider 
iTJ as Aramaic ; this is philologically correct. But for Kalir and 
Saadiah, the word was pure Hebrew, because, according to their 
grammatical views, that the J"S verb JU3, of which the Bible has 
the forms fW (2 Sam. xxii. 29) and >\V (Job xviii. 5), »1J only could 
be considered as the root, and they could form n^ from TO, analogous 
to D*S (in their opinion formed from riS) and Vfo (in their opinion 
formed from nfe>). 

Prof. Margoliouth proceeds: "The word DB>J is exceedingly faulty 
for flDEO." The reply to this is, that according to my remarks 
(I.e., pp. 23, 181, 188-9, l 9 2 > 2 38) the use of the incriminated word 
is fully justified. I showed there that, according to Saadiah's own 
evidence, it was a universal poetical custom in those days to trans- 
form feminine into masculine words, that Saadiah himself had, most 
probably in another fragment of Ha-Galuy, transformed fUJjp into }$£>, 
and that Saadiah's pattern, Kalir, also used the word DEO . All this 
is entirely disregarded by Prof. Margoliouth, although, I think, that 
it is, at least, worthy to be refuted. I can now add, that many 
similar words used by Saadiah belong to the same category ; e. g. 
D1?nN, quoted in Dunash's Critique, from the Biblical iTO?nN, fUN, 
in his may (p. 12), from 3in nrUN (Ezek. xxi. 20), epB (ibid., p. 13) 
from the Talmudical word naiJO, Aramaic NriBUQ, "ID? (p. 14) from 
mot, 3HN (p. 15) from iUriK, &c, &c. But the matter is clinched 
by this, that already Zunz (ibid., p. 384) cited the corpus delicti, the 
word DB>3, from a liturgical poem of Saadiah's, commencing with 
the word N">pn (Saadiah's dirge wp V\W tOpn is probably meant, 
vid. Literaturgeschichte der synagog. Poesie, p. 97). 

Prof. Margoliouth says further : " The form TOT^O is not Biblical, 
but Arabic." This cannot be asserted so offhand, for we have already 


in the second verse of Genesis the form nBCPD, cf. mtJ'SD (Isa. xl. 9), 
ni3HD (1 Sain. i. 13), ftrDTD, plur. (1 Kings xi. 8), &c, &c. Saadiah 
has himself mBDD in JTOK (I.e., p. 54, ver. 12). Prof. Margoliouth 
probably meant to say that a verb was formed here from the noun 
3I"6, but such mode of proceeding is sufficiently known to us from 
Saadiah's poetical productions. I quote a few examples : In the 
rnuj) he has bwn from B^m, W, i»e>Nn from bf» (p. 10), 
W>trt from nb%, noblB from the Talmudical KD^IB (p. 12), O'Oin 
from DCn (Jer. xlix. 24), BbJ from e*?^ (Ezek. xiii. 11, p. 13), &c. 
In a fragment quoted by me (p. 189) Dnbnn from DrO (Esther i. 6). 
Most instructive are in this respect the instances quoted by Dunash 
in his Critique (No. 88), for we learn from them, that the Gaon 
formed JirfiK from rfN, nnft from m, and even transformed proper 
nouns like VSp, and interjections like T1VJ and nntf into verbs ! 

The Hebrew and Arabic titles ^Jil 1BD and' "ONt^X awofo 
present difficulties to Prof. Margoliouth, as they did to other critics 
before him, but without reason. I have already given the only correct 
explanation (I.e., pp. 142, 180), namely that the words mean "the 
Book of the Exiled " (of the exiled one), I only forgot to add that we 
ought to read in the Arabic "P1D7N 38*13 (or also TIDPN, in a 
passive sense), just as in the Hebrew the word *1?J is taken as the 
passive of n?i (to exile). Thus he has also in the other fragment(p. 189) 
'13n, analogous to the Biblical , fln, and MD3 in Ben Sira. After my 
above remarks about Saadiah's artificial style, Prof. Margoliouth will, 
I hope, admit that my explanation does not prove at all that " Harkavy 
strangely prefers the barbarous Hebrew to the correct," &c, but that 
every author must be judged according to the conditions of his age, 
according to the linguistic stage of his period, and, principally, 
according to his own works and mannerisms. 

12. Having thus shown how groundless Prof. Margoliouth's reasons 
are for suspecting the work, it will be sufficient only briefly to refer 
to the remaining points of his criticism, which rest either upon 
premises, which can be proved to be incorrect, or upon arbitrary 
assumptions, and I shall only enter upon a detailed discussion of the 
following remarks should the specialists on these subjects find that 
my defence of the Ha-Galuy fragment is faulty, and that the genuine- 
ness of the book is doubtful. 

(a) Saadiah's opponent Mubashir Halevy, who wrote after Saadiah's 
death (942), as noted by me in the first part (p. 68), cannot, of 
course, be identical with the Gaon Mubashir, as Prof. Margoliouth 
thinks. The latter Gaon was no Levite, and was already dead in 926. 

(&) Saadiah's work iTDflfl 1BD, quoted by Abraham bar Chija in 


the "lttyn 1BD, was recognized long ago to be the POWK 3W13, 
which was written in Arabic, and directed against the Karaites. 
It is the same book as the one entitled )ri3Dn "I2D, in the translation 
of Moses ibn Ezra's fipnn^K ff?HPD. 

(c) Saadiah, like the Arabic authors, understood under Nabataean 
the Arabaic mixed with Aramaic and Persian words, which was 
spoken in Babylonia. 

(d) Prof. Margoliouth's supposition that the Galuy fragment is 
a complete composition is obviously erroneous. Both its external 
shape and its contents (it commences in the middle of a phrase with 
the vav conjunctive) prove clearly that it is d*«$aXoi> xdi ariktm-ov. 
The supposition is also refuted by the quotations from the work : 

innj?D i>rm (p. 163), \rmvD rvn nx 'n -ijm (p. 165), nosnoi, lrmyn 
anoc ( P . 167), Dwmn ijv on^, nuji ptn Dinroi (p. 161). 

(e) Neither is the MS. whole, nor is it an autograph from the year 
962, but it most likely belonged to a copyist of the twelfth century, 
who has not copied everything correctly. 

(/) Prof. Margoliouth's positive assertion that Saadiah's book of 
the festivals was composed in Arabic, will hardly impress anybody, 
unable as he is to give a single quotation from it. He does not take 
any notice of the fragment printed by me (1. c, p. 220). 

(g) Prof. Margoliouth's fantastical combination, about a Karaite 
fabrication of the Galuy fragment, is hardly worth a serious refutation. 
Not even the cleverest forger could have fabricated a literary pro- 
duction composed in Saadiah's style, written in words peculiar to 
Saadiah, and so thoroughly impregnated with the ideas and the 
spirit of Saadiah, let alone a Karaite of the tenth century. 

In every Karaite forgery, both old and new, a great number of 
which the writer of these remarks has first recognized and pointed 
out, it is always an easy matter to find out the cui prodest, and the 
Karaite character is always more or less glaringly conspicuous. 
Where are the signs of Karaism in our fragment ? Should its general 
object, the ridiculing of Saadiah, be its caricature ? But it contains 
nothing which does not correspond with the Gaon's position at the 
time, or which could not per se be proved, from other works, to be his 
property. It would indeed be a rare curiosity in the history of Jewish 
literature, such a Karaite satire upon Saadiah, which was never used 
by a single one of the Gaon's many Karaite antagonists, but which 
was naively used by Nissim of Kairouan, a zealous Rabbinite of the 
first half of the eleventh century, who had constant intercourse with 
Babylonia, and who wrote polemical writings against the Karaites 
(vid. my remarks on this in Steinschneider's Festschrift). It would 
be a rarity far beyond anything hitherto considered rare. 


(ft) The unjust proceedings, which according to Nathan Babli's 
report, were the cause of the strife between Saadiah and the Exilarch, 
can only explain the tatter's hatred against the former, but not the 
antagonism of the academy of Pumbeditha, of the Gaon Cohen Zedek, 
of the later Gaon Aaron (Chalaf) ibn Sarjado, and of all other 
opponents of Saadiah. We learn from the Galuy fragment — what 
the introduction to the Amanath already made appear probable — 
that Saadiah's learned aspiration was also one of the causes. There 
is, therefore, no contradiction between Nathan's report and the 
preface of Ha-Galuy. For the rest, the expression TTJJ V&\ (and no 
justice) in the preface shows quite clearly, that justice had been 
infringed by Saadiah's opponents, and this subject was of course 
fully dealt with in the work Ha-Galuy itself. 

(t) Prof. Margoliouth has overlooked the fact that the title of 2NTON 
?1J??N1 TronpN (people of monotheism and justice) was in reality the 
title borne by the Arabic Mutakalemin, and that it was appropriated by 
Mutakalemite Karaites from a love of imitation. This was known since 
Delitzsch (1841). Cf. also Schreiner, DerKalam (Berlin, 1895, p. 5). 

(k) Prof. Margoliouth disregarded also my remark about the title 
of P313T (p. 153, note 10). There the matter is quite satisfactorily 
explained. Namely, Saadiah launches against his opponents Cohen 
Zedek, Sarjado, &c, the reproach, that they had no proper historical 
knowledge of Rabbinism, notwithstanding the fact that they now 
were called Rabbis and constantly had that title in their mouths. 
This is, therefore, directed against the present Rabbis, i.e. against 
Saadiah's opponents, and not against the Rabbis in general. How 
is it possible to recognize here, with Prof. Margoliouth, " the Karaite 
hand," and how is here " Saadiah's own party made ridiculous " ? 

(I) Prof. Margoliouth, in discussing David ben Zakkai's claims to 
the position of Exilarch, forgot that the latter had been first deprived 
of his office by Saadiah, and superseded by his brother, and that only 
afterwards David's party gained power, of course, by means of bribing 
the Mahommedan authorities. Consequently, Saadiah could justly 
consider Ben Zakkai as an illegitimate Exilarch. 

(m) Prof. Margoliouth thinks it to be "impossible" for Saadiah 
to speak boastingly of himself, but he disregards the distinct evidence 
of Ibn Daud (in Neubauer's Mediaev. Chron., I, 66), who says: ""IJV1 

^v\ ibd bv D^ina oan bvrwb rwv ipn rroiDm nnyD n nan." 

It may be that, according to our present ideas, one or another boast- 
ful expression may appear improper ; but when we take into 
consideration the time when it was written, the literary fashion of 
the Arabic writers of the age, and the personal position of the Gaon, 
we shall become more lenient in our judgment. 


(») The readers of Saadiah' s writings were long aware of the fact 
that he repeatedly dealt with many subjects in his various works. 
Abraham bar Chija gives distinct evidence (cf. my work, pp. 133, 143, 
155) that Saadiah fully discussed the time of Israel's deliverance 
(the fP) both in Ha-Gdluy and in the Amanath. Most probably he 
considered it necessary and useful to treat this subject in a pamphlet 
also, after having discussed it in his large work on philosophy of 
religion, which was only written for scholars. I have a conjecture 
about this, but it would lead me too far here to enter upon it. At 
any rate, it is unintelligible why the time of the deliverance when 
fixed by Saadiah as 933 may have been meant seriously, but when fixed 
as 934, it must, in spite of Abraham B. Chija's evidence, he put off 
to 962, and be meant satirically. 

(0) It is true that Saadiah had done all he could to avoid the 
strife ; but it became inevitable when it transpired that the Exilarch 
would not yield anything of his criminal demands, and made use of 
violent measures. On what compromise with the Exilarch could 
the Gaon enter after this ? 

(p) Everybody will think it only natural that Saadiah 's polemic in 
the book Amanath and elsewhere is calmer, milder, and more modest 
than in Ha-Galuy. In the former, the polemic is not, as in the last, 
directed against personal enemies, who had taken everything away 
from him, who had personally illtreated him, had imperilled his life, 
and had publicly boasted that they had acted thus. And since Saadiah 
was usually calm and moderate, we can only conclude, that Ha-Galuy 
was a book written with a purpose (in self-defence against personal 
enemies), which if not justifiable, was at least excusable, in view 
of the conditions of the time. 

The foregoing remarks are, I think, sufficient to set aside Prof. 
Margoliouth's principal strictures upon the Galuy fragment. Should 
it, however, be deemed necessary, I am prepared to disprove many 
other alleged proofs of Prof. Margolionth. I only want to draw 
attention to this, that the assumption, that we have to deal here 
with a Karaite satire, entails greater difficulties than Prof. Margoliouth 
believes he has found. How is it possible that a Karaite satire 
against Saadiah existed since 962 without a Karaite having made use 
of it ? That the famous head of a school in Kairouan of the eleventh 
century should have taken it, without any misgivings, for a genuine 
work, and should have borrowed from it forged verses of Ben Sira, 
and even Menasseh's argument, so strongly censured by Prof. 
Margoliouth ! And we are to believe that this Karaite forger of 
the tenth century undertook the difficult task, without considering 
that any Rabbinite could unmask and disgrace him by producing the 


genuine Galuy\ Prof. Margoliouth may pardon me — but were it 
not that his esteemed name appeared at the bottom of the article, 
and if the latter did not contain some side issues, discussed with 
great erudition and acumen, it might be more reasonably taken as 
a satire against many a modern critic (especially in the field of Bible 
criticism), rather than the Galuy fragment as a satire against Saadiah. 

A. Harkavy. 
St. Petersburg, January, 1900. 


The editors of the Cambridge fragments of the Hebrew text of 
Ecclesiasticus (1899) discuss the best way of filling up S . . K . . .3TO ^33 , 
and suggest (p. xlvii) the reading !»DN)1 *IBn an? ^33 (or ^33). 
M. Hal^vy (he nouveau fragment hibreu, p. 4) suggests as the last word 
*)D33. I venture to express the opinion that 7*083 is evidently a 
corruption of ?t3"UK or T'B'UK. 7CH28 occurs in Ezra i. 9, where the 
versions (LXX, Vg., Syr.) give the sense "bowl" or "bason." The 
English Version gives " bowls of gold, silver bowls." Should we not 
read (correcting and transposing) *|D3 ?t3">!W ? So the sense becomes, 
" A vessel of gold and (or) a bason of silver." 

March 15, 1900. T. K. Cheyne.