1

The Fifteenth International Orion Symposium, in conjunction with the University of Vienna, Institute for Jewish Studies, and the Schechter
Institute of Jewish Studies
April 1014, 2016

The Karaites and the Hebrew Bible

Geoffrey Khan, University of Cambridge
Karaism arose in the early Islamic period as a movement within Judaism whose
essential difference from Rabbanite Judaism consisted in their rejection of the legal
authority of the oral law embodied in the Mishnah and Talmuds. The Karaites held
that the basis of legal authority was the Bible. They shared the Bible in common
with Rabbanite Jews and felt no need to create a separate version of Scripture that
deviated from that of the Rabbanite Jewish tradition. This should be contrasted with
the Samaritans, who broke away from Judaism and expressed their communal
identity by adopting a different tradition of Scripture with its own distinct
modifications and tendentious additions. The fact that the Karaites and the
Rabbanites shared the same scripture meant that Karaites and Rabbanites had a
common interest in carefully preserving the transmission of this scripture.
In the Middle ages all Jewish communities shared the same consonantal text of
the Bible, but there were conspicuous differences across various communities
regarding the reading traditions of the text, i.e. the ways in which the text was read
aloud. This is reflected in the different vocalization sign systems that are attested in
medieval manuscripts. These sign systems can be classified broadly into the
Tiberian, Palestinian and Babylonian systems, though in each case the manuscripts
attest to a number of variant sub-systems. The different vocalization system do not
represent any substantial differences in text among the different reading traditions,
but they do reflect considerable differences in pronunciation and also, in many
cases, differences in morphology.
The so-called standard Tiberian system of vocalization signs was regarded in
the Middle Ages as the most prestigious and authoritative system. The reading
tradition that the Tiberian vocalization reflected, known as the Tiberian reading
tradition, was held to have similar prestige. The Tiberian vocalization sign system
was developed in the early Islamic period by the Masoretes of Tiberias. The purpose
of the sign system was to function as a written notation representing a tradition of
reading that had been transmitted orally over many generations since the Second
Temple Period. The activities of the Tiberian Masoretes came to an end in the 10th
century. Shortly thereafter the Tiberian reading tradition fell into oblivion among
Jewish communities. The Tiberian sign system, however, which had been developed
to represent the reading tradition was eventually adopted by all Jewish communities
and it replaced the other sign systems. As a result the Tiberian signs came to be read
with various local reading traditions, none of which were direct continuations of the
Tiberian reading tradition.

It is important to note that the Karaites shared with the Rabbanites not only a
common scripture but also shared with them the recognition of the Tiberian
tradition of scripture as the most authoritative. It is for this reason that the Karaites
had a particular interest in the Tiberian Masoretic tradition.
Karaites were closely associated with the Tiberian Masoretic tradition. This is
demonstrated in a number of ways.
Many of the surviving monumental Tiberian Masoretic codices that were
written towards the end of the Masoretic period in the 10th and early 11th centuries
contain inscriptions that indicate that the manuscript was dedicated to a Karaite
community.1
Bible manuscripts often passed from Rabbanite into Karaite hands and vice
versa and some of the manuscripts dedicated to Karaite communities may have
been originally in the possession of Rabbanites. Such is the case with the Aleppo
codex. This manuscript contained an inscription that indicates that it was written
by the scribe Shlomo ben Buy and its vocalization and Masora were supplied by
Aharon ben Asher and that a certain Israel ben Sima of Bara dedicated the
manuscript to the Karaite community of Jerusalem. The first folio of the manuscript,
however, contained a notice recording the fact that it was dedicated to the
Rabbanite community of Jerusalem () .2 The
inscription indicating that the manuscript was dedicated to the Karaite community
should be dated to the middle of the 11th century, about a hundred years after the
manuscript was first produced by Ibn Buy and Ben Asher. The notice at the
beginning of the manuscript suggests that it was originally in Rabbanite hands and
was subsequently purchased by the Karaite Israel ben Sima. This manuscript was
transferred to Egypt after the capture of Jerusalem by the Crusaders, where it was
placed in the Rabbanite synagogue of the Palestinians. When in Egypt, it was
consulted by Maimonides, who pronounced it to be the most reliable model
manuscript.3 By the 16th century the manuscript had passed into the possession of
the Rabbanite community of Aleppo.4
1 Evidence of such dedications are found in the colophons of several of the Bible manuscripts in the

second Firkovich collection published by Kahle (1927, 5677), e.g. no. 2 Cod. 159, dedicated to the
Jerusalem Karaite community, 937 A.D. , no. 3 Cod. 10, dedicated to the Fustat Karaite community,
eleventh century , no. 8 Cod. 223, 225, dedicated to the Jerusalem Karaite community, 1017 A.D. , no. 11
Cod. 25, 26, dedicated to the Fustat Karaite community, eleventh century , no. 12 Cod. 94, dedicated to
the Karaites of Egypt, 1100 A.D. , no. 13 Cod. 34, dedicated to the Karaites of Fustat after the death of its
owner, eleventh century . See also the colophons published by Poznanski (1913), e.g. p. 115 dedicated to
the Karaites of Ramla, 1013 A.D. .
2 The original inscriptions are now lost and survive only in copies Kahle 1927, 712; Ofer 1989 .
3 See Goshen-Gottstein (1960), Penkower (1981) and Ofer (1989) for the evidence that the Aleppo

Codex was indeed the manuscript that Maimonides saw in Egypt. It is generally believed that this
pronouncement of Maimonides ensured that the Ben Asher Masoretic tradition became the

Some Tiberian Masoretic Bible codices that have inscriptions indicating their
dedication to the Karaite community of Jerusalem in the 10th or 11th century and were
subsequently transferred to Egypt were held in the possession of Karaite community
of Cairo down to the present. This applies to several of the old Bible codices kept in
the Karaite synagogue of Cairo (Gottheil 1905, 647), including the manuscript
known as C3 (10th century, no. 18 in Gottheils list) (Penkower 1989), and the socalled Cairo Codex of the Prophets (datable to the 11th century, the colophon
indicating the date 895 C.E. being a copy of an earlier colophon).5
Although a dedication of a manuscript to a Karaite community does not
necessarily mean that it originated in Karaite circles, the identification of the person
who commissions the writing of the manuscript as a Karaite is proof that its
production was a Karaite initiative. This is, indeed, the case with the codex
Leningradensis (I Firkovitch B19a), which was commissioned by the Karaite
Mevorakh ben Joseph ha-Kohen, who was, therefore, its first owner. According to
the colophon of this manuscript it was written by the scribe Samuel ben Jacob in the
first decade of the 11th century (the various dating formulas correspond to dates
between 1008 and 1010). Samuel is said to have added the vocalization, accents and
masoretic notes based on the corrected and clear6 books of the teacher Aharon ben
Moshe ben Asher, may he rest in Eden (

) . This, therefore, was a copy of a Tiberian Masoretic
manuscript and was not itself produced by a Tiberian Masorete. Jacob ben Samuel
also wrote a Bible manuscript preserved in the Karaite synagogue in Cairo, no. 14 in
the list of Gottheil (1905). The colophon indicates that this was commissioned by
David ben Yeshua ha-Levi who presented it to the Karaite community of Cairo. So
the production of this manuscript, too, appears to have been a completely Karaite
initiative.7

authoritative one in Judaism. A source from the eleventh century refers to the possibility of
following either the school of Ben Asher or that of Ben Naphtali, without any evaluation (Eldar
1980). It is relevant to note, however, that the original vocalization and accents of the manuscript
C3 of the Karaite synagogue in Cairo (Gottheil 1905 no. 18) exhibited features of Ben Naphtalis
system but these features were corrected by Mishael ben Uzziel in the early 11th century to the
reading of Ben Asher (Penkower 1989).

(1927, 312) and BenZvi 1960 .

5 For the arguments regarding its dating, see Cohen (1982), Glatzer (1989, 25059), Lipschtz (1964,
67).
4 See the reconstruction of the history of the manuscript by Kahle

6 Kahle interprets the term as supplied with Masora.

7 Jacob ben Samuel also wrote the Bible manuscript no.27 of Gottheils list, which was preserved in the
Karaite synagogue, for a certain Yaya ben Jacob.

The dedication inscription of the Aleppo codex indicates that even when this
carefully produced model manuscript that was written by Tiberian Masoretes came
into the possession of a Karaite community, it was made available also for
Rabbanites to consult in order to check readings:

If anybody of the seed of Israel,

from among the Karaites or the Rabbanites, wishes on any day of the year to see in
matters relating to full or defective orthography, what is disordered or ordered,
closed or open sections, or one of the accents, they should bring it out for him to see
and check and so gain understanding (Kahle 1927, 45; Ofer 1989, 28889).
Some scholars who are known to have been Karaites can be shown to have
been closely associated with the Tiberian Masoretic tradition. This applies to the
Karaite grammarians who were active in Jerusalem towards the end of the
Masoretic period. The two most important Karaite grammarians in this repect are
Ab Yaqb Ysuf ibn N (known in Hebrew as Joseph ben Noa, second half the
tenth century) and Ab al-Faraj Hrn ibn Faraj (first half of the eleventh century).8
According to Ibn al-Ht, who wrote a chronicle of Karaite scholars, Ysuf ibn
N had a college (dr li-l-ilm) in Jerusalem, which appears to have been
established around the end of the tenth century.9 Ab al-Faraj Hrn belonged to
this college of Karaite scholars. Some sources refer to Ysuf ibn N Ab Yaqb
Ysuf ibn Batawaih (or Batawi), or Joseph ben Batawaih. Batawaih may have
been the Iranian equivalent of the name N or Noa (cf. Persian bat fortune,
prosperity). Inscriptions on some of the early Masoretic Bible codices that dedicate
the manuscript to the Jerusalem Karaite community give instructions for them to be
deposited in the aer (compound) of Joseph ben Batawaih, which is likely to be
identical with Ibn Ns college, referred to by Ibn al-Ht by the corresponding
Arabic term dr.10 The manuscript C3 of the Karaite synagogue in Cairo (Gottheil
1905 no. 18; Penkower 1989) contains the inscription
I Mishael ben Uzziel ben Yoseph ben
Hillel checked this holy Torah in the enclosure of ben Bakhtawaih, may God have
8 For details of the Karaite grammarians of Jerusalem and their grammatical thought see Khan 2003;
2014a; 2000 , Khan, Gallego and OlszowySchlanger 2003 , Vidro 2009; 2011 .
9
. For the text of Ibn al-Ht see G. Margoliouth (1897: 433, 438-39). Ibn al-Ht was writing in the fifteenth
century. For the background of Ibn Ns college, see J. Mann (1935: 33-34).
10 Cod. 159 of II Firkovitch collection, dedicated 937 A.D. Kahle 1927 no.2, p.60 , Cod. 225 of II Firkovitch
collection, dedicated 1017 A.D Kahle 1927 no. 8, p.67 . The inscription in cod. 159 relating to the
compound of Joseph ben Bakhtawaih is separate from the main dedication, which is dated 937 A.D. Since
Ysuf ibn N/Joseph ben Bakhtawaih is known to be active at the second half rather than the first half
the 10th century, the inscription relating to his compound must have been added later. The inscriptions in
both manuscripts add the blessing may God have mercy on him after the name Joseph ben
Bakhtawaih, which indicates that they were written after his death.

mercy on him. This indicates that it must have been kept in ben Bakhtawaihs
enclosure, where the scholar Mishael ben Uzziel consulted it.
The close relationship of the early generation of Karaite grammarians in the
tenth century, such as Ibn N, to the Tiberian Masoretic tradition is reflected by
the methodology and disciplinary structure of their works. The grammatical work of
Ibn N that has come down to us is a grammatical commentary on the Bible
known as the Diqduq. The term diqduq does not have the sense here of grammar as
an independent discipline, but rather as the study of the fine details of scripture. A
central feature of Ibn Ns method of presentation is the explanation as to why a
word has one particular form rather than another. This often involves comparing
closely related forms that differ from the form that is under investgation only in
small details. The issue that is addressed is why these fine distinctions in form exist,
with a view to finding hermeneutical significance in them. This may be compared to
the practice of the Masoretes to collate words that were similar in form but differed
only in details. This was a central feature of the masoretic method and lists
recording these collations are found throughout the masoretic notes that were
attached to Bible codices. The purpose of this was to draw attention to fine details of
form to ensure that they were preserved in the transmission of Scripture. Collations
of two closely related forms of word were also compiled in independent masoretic
treatises, such as Olah we-Olah.11 By the tenth century, the Masoretes also
compiled treatises that formulated rules for the occurrence of some of these fine
distinctions in form with regard to vowels and accents. The most famous work of
this kind is the Diqduqe ha-eamim The rules of the details of the accents, which
was compiled in the first half of the tenth century by Aharon ben Asher.12
Apart from these parallels in methodology, another feature that reflects the
close relationship of Ibn Ns work to Masoretic activities is the fact that Ibn Ns
grammatical comments relate mainly to morphology, with some consideration of
syntactic issues. They make only very marginal reference to issues of vocalization
and accents. They, therefore, complement the Masoretic treatises, indicating that
the early Karaite grammarians were developing the Masoretic tradition rather than
setting up an independent discipline.
This complementary relationship between the Masorah and grammar which
characterizes the early Karaite grammatical tradition contrasts with the scope of the
grammatical work composed by Saadya Gaon in the tenth century. After leaving
Egypt, Saadya spent a few years in Tiberias studying among the Masoretes.
11

. The treatise Olah we-Olah is named after the first two words of the first list (eating [I Sam. 1:9] and
eat [Gen. 27:19]), which enumerates pairs of words, one occurring with the conjunctive waw and the other
without it. For a general discussion of the background of the text see Yeivin (1980: 128-131). An edition of the
text based on the best manuscripts has been made by Daz Esteban (1975) and Ognibeni (1995).
12
. The definitive edition of this text is by Dotan (1967).

According to Dotan (1997, 3340) he composed his grammar book (Kitb Fas
Luat al-Ibrniyyn The Book of the Eloquence of the Language of the Hebrews)
while he was in Tiberias during the second decade of the tenth century. Unlike the
Diqduq of Ibn N, the grammatical work of Saadya does not complement the work
of the Masoretes but rather incoporates numerous elements from it. The surviving
sections of the work include not only treatments of grammatical inflection and word
structure but also several chapters relating to the Tiberian reading tradition. The
material for some of these has clearly been incorporated from the Masoretic
tradition and direct parallels can be found in the extant Masoretic treatises such as
Diqduqe ha-amim (Dotan 1997, 3436). Saadya refers to the accents on various
occasions. Dotan, indeed, suggests that one of the missing chapters from Saadyas
work may have been concerned specifically with accents. We may say that Saadyas
grammar book is not a product of collaboration with the Masoretes or a
complementary expansion of the scope of Masoretic teaching, as is the case with the
Diqdquq of Ibn Nh, but rather was intended as standing separate from the
Masoretic tradition.
The complementary relationship of Ibn Ns grammatical work with
Masoretic activity is further shown by an early text published by Allony (1964) that
contains a list of technical terms for the various aspects of Biblical study. These are
described in the text as diqduqe ha-miqra, which has the sense of the fine points of
Scripture established by detailed investigation. The list includes masoretic,
grammatical and hermeneutical terms. These correspond closely to the terminology
and concepts of Ibn Ns Diqduq. The range of the topics of analysis denoted by the
terms also parallels the scope of analysis that is found in the Diqduq, though, as we
have remarked, the focus of the Diqduq is more on the grammatical and
hermeneutical aspects than on the masoretic. It is more accurate to say that the
masoretic works and Ibn Ns Diqduq combined cover the range of topics
contained in the list. The Masora and Masoretic treaties, on the one hand, and the
grammatical work of Ibn N, on the other, complement each other to establish the
diqduqe ha-miqra. This list was not intended primarily as a foundation for the study
of grammar per se, but rather as a methodology for establishing the correct
interpretation of Scripture. It appears to reflect the scope of the scholarly work that
developed around the core Masoretic activities, in which the grammatical work of
Karaite scholars such as Ibn N play an integral role.
Allony, in his edition of this list of technical terms, claimed that it was of
Karaite background. One should be cautious, however, of being too categorical on
this issue. Certain details of its content suggest that it was composed in the early
Islamic period. It would, therefore, come from a period when Karaism was in its
embryonic stages of development. The main evidence that Allony cites for its being

a Karaite work is the reference in the text to the masters of Bible study (baale hamiqra). This term was used in some texts in the Middle Ages to designate Karaites.13
It is found, however, already in Rabbinic literature in the sense of those who study
only the Bible and not the Mishnah or Gemara.14 It should be noted, moreover, that
in masoretic texts it is sometimes used as an epithet of the Masoretes, who were
professionally occupied with the investigation of the Bible.15 The contents of the list
were incorporated by a number of later authors into their works. These included not
only Karaites but also Rabbanites, such as Duna ben Labra.16
Ab al-Faraj Hrn, who was based at the college of Karaite scholars, is said by
Ibn al-Ht to have been the student of Ibn N. The content of Ab al-Farajs work,
however, is very different from that of Ibn N, mainly because it conforms more
closely to the theories of grammar propounded by the mainstream Baran school of
Arabic grammar. The Arabic works of Ab al-Faraj, nevertheless, still maintain the
complementarity between grammar and Masora, which is charateristic of the work
of Ibn N, in that they are by and large restricted to morphology and syntax with
minimal consideration of vocalization and accents, which reflects the close
relationship of Ab al-Faraj to the Masoretic tradition, again contrasting with the
grammatical work of Saadya.
Moving now from the para-masoretic discipline of Karaite grammar to core
masoretic activities, we are able to identify the authors of some of Masoretic
treatises at the end of the Masoretic period in the 11th century as Karaite. Ab alFaraj Hrn himself wrote a treatise on the biblical reading known as Hidyat alQri Guide for the Reader, which described the pronunciation of consonants and
vowels and the principles of the accents (Eldar 1994). This complemented his
grammatical works on morphology and syntax (Khan 2014b). He states that his
sources for the work were earlier masoretic treatises and the pupils of the writers of
these earlier treatises (),17 which indicates that he had access to an oral
tradition of instruction in the Tiberian reading. Another scholar working in the
Karaite college founded by Ysuf ibn N/Joseph ben Batawaih was Mishael ben
Uzziel. As we have seen above, in an inscription in the manuscript C3 of the
Karaite synagogue in Cairo he writes that he checked the manuscript in the
enclosure of ben Batawaih. This implies that he was a Karaite belonging to the
circle scholars in the Karaite college, most likely contemporary with Ab al-Faraj
Hrn in the first half of the 11th century. This is almost certainly the the same
13

. It is used frequently in this way in the inscriptions on the Bible codices discussed above and also in the
writings of medieval Karaite scholars such as Salmon ben Yeruam and Judah Hadassi.
14
. Cf. Bacher (1899: 118).
15
. E.g. Baer and Strack (1879: xxxviii).
16
. Teshubot de Dunash ben Labrat, ed. A. Saenz-Badillos, Granada, 1980, 15*.
17 MS Evr. Arab. I 2390, fol. 7a.

Mishael ben Uzziel who composed a Masoretic treatise concerning the diffferences
between the leading Masoretes Aharon ben Asher and Moshe ben Naphtali, known
as Kitb al-K ilaf The book of differences (Lipschtz 1964; 1965). Fragments are
extant also of a book of differences between the Masoretes written at roughly the
same period by the Karaite Levi ben Yefet, also know as Levi ben al-H asan, who was
the son of the Karaite translator and exegete Yefet ben Eli (Lipschtz 1964, 3).
Several modern scholars have argued that some of the Masoretes themselves
were Karaites, in particular Aharon ben Asher, who was one of the most prominent
Masoretes towards the end of the Masoretic period in the 10th century. In all cases,
however, the arguments are based on indirect or doubtful evidence.
Pinsker (1860, 34) says that all the Masoretes should be suspected of being
Karaites since they spent their time occuped with vocalization and accents of the
Bible and there is no evidence that they showed any interest in the Talmud.
Klar (1954, 276319) identified the name Ben Asher in a manuscript at the
head of Saadya Gaons polemical poem against the Karaites , and claimed
that this proved that Saadyas Karaite opponent, against whom the poem was
addressed, must have been the Masorete Aharon ben Asher. Zucker (1958) and
Dotan (1977) have convincingly argued against this on the grounds of the contents
of poem, which are inconsistent with such an identification, especially the
attribution of anti-Talmudic pronouncements to the opponent.
Some scholars (e.g. Graetz 1881, 366; Klar 1954, 276319) have claimed to be
able to identify Karaite doctrines in Masoretic treatises attributed to Aharon ben
Asher and in the extant poem of the vine that is attributed to his father, Moshe ben
Asher, but again these are not at all clear and have also been rebutted by Zucker
(1958) and Dotan (1977). More recently Zer (2003) has argued that evidence for the
Karaite persuasion of Aharon ben Asher can be found in the masoretic notes that he
wrote in the Aleppo codex. One of Zers central arguments is based on the masoretic


( Job 8:8), which reads:
note on

It (i.e. the word
) is defective (in orthography) unlike

any (other occurrence of the word) in the Bible. Why? Because in the first
generation all the commandments that are in the Torah were not completed, but it
lacked many commandments, therefore it (i.e. the word
) is unique in the Bible,
because the commandments were only completed by the hand of Moshe, our
master. Zer points out that Karaites held this doctrine of the gradual revelation of
commandments. Erder (1994) has presented various medieval Karaite sources that
adhere to this doctrine. There is, however, some degree of variation of thought in
these sources. Moreover the view that commandments were given before Sinai is

also found in the Talmud and in the writings of Maimonides, although Maimonides
maintains that only those revealed at Sinai are obligatory (Erder 1994, 13739). The
use of differences in full and defective orthography of the ketiv as a source of
interpretation is a practice found in Rabbinic texts (Goldberg 1990), but it was not
approved of by many medieval Karaites, who regarded the reading tradition (qere)
to be the only legitimate source of textual authority. This is stated explicitly by
al-Qirqisn.18 It would, therefore, be highly unusual for a Karaite Masorete to use
this type of hermeneutics.
Yeivin (1962) draws attention to a peculiarity in the vocalization of the Aleppo
codex whereby in words that have a qere with inversion of the letters of the ketiv, the
vocalization signs are not marked in the order required by the qere but rather are
marked on the letters of the ketiv in a different order from the word of the qere that
they are intended to represent. In Ezek. 36:14, for example, the ketiv has a qere
with inversion of letters, which the codex Leningradensis represents , i.e. the
vocalization reflects the form you will [not] bereave with the vowels in the
correct order. The Aleppo codex, however, represents this , with the vowels of
the qere placed directly under the consonants of the form of the qere, with the result
that the vocalization does not follow the order of vowels in the qere. Another
example, this time a word with an accent, is in 2 Sam. 20:14, where the ketiv is
represented by L , reflecting the form and the assembled, but the Aleppo
codex vocalizes: . Yeivin proposes that this unusual practice of vocalization in
the Aleppo codex reflects the view of the vocalizer that the vocalization and accents
were given at Sinai and are as old as the letters, with the consequence that the vowel
signs were considered to be bound to the letters, i.e. if the letters are inverted, then
the vocalization and accents were also inverted. Yeivin argues that this demostrates
that the vocalizer, i.e. Aharon ben Asher, was a Karaite, because the Karaites held
the view that the vowels and the accents were given at Sinai. There is not, however,
clear evidence of such a doctrine being held by the Karaites, to my knowledge,
before Judah Hadassi (12th century), who states in his work Ekol ha-Kofer that that
the original Tablets given to Moses at Sinai had the vowels and accents for without
the five vowels which are [represented by] the vowel signs a word could not be
articulated nor could it be understood without the pronunciation of the vowels and
accents.19 Earlier Karaites, however, did not express this doctrine. The Karaite list of
grammatical terms published by Allony (1964) and dated by him to the Masoretic
18 The relative passages of alQirqisns work are discussed below.
19

( Hadassi 1836, 70a).

10

period only states letters are [known] by vowels

and the vowels are known by the masorot. One cannot infer from this, as Allony
does (1964, 333, 349), that the author believes that the original Tablets were
inscribed with the vowels and accents. The Karaite al-Qirqisn (10th century) states
that the vowels are subordinate to the letters and are not part of the holy script.20
At approximately the same period as Hadassi expressed the doctrine that the vowel
points were given to Moses at Sinai, the same view was expressed by several
Rabbanites, e.g. David Qimi (Sefer Milol, ed. Lyck, 1862, 73) and Moses ben Isaac of
England.21 There is no evidence that the doctrine was specifically Karaite. It seems to
have arisen due to the chronological distance between the scholar in question and
the period in which the pointing and accentuation were first developed rather than
due to any doctrinal Tendenz.
We may conclude, therefore, that there is no incontrovertible evidence that
Aharon ben Asher or his family were Karaites. The medieval sources refer to several
generations of Masoretes, some of them belonging to the same family. They indicate
that the family of Aharon Ben Asher had been involved in masoretic activies over
five generations. Aharon Ben Asher lived in the 10th century and so Asher the elder,
who is stated to be the great-great-grandfather of Aharon, is likely to have lived in
the second half of the eighth century C.E., before the emergence of Karaism on the
historical scene in Palestine (Kahle 1959: 75-82; 1927, vol. 1, 39). Some of the
Masoretes, furthermore, were closely associated with the Rabbanite Jewish
authorities, e.g. Pinas Rosh ha-Yeshiva (head of the academy), who lived in the
ninth century, and Aiyyahu ha-Kohen he-Haver (member of the academy).22 The
academy (yeshiva) was the central body of Rabbanite Jewish communal authority
in Palestine. Some close parallels to the format and phraseology of the masoretic
notes can, in fact, be found in Midrashic literature composed before the Islamic
period (Martin Contreras 1999, 2002, 2003). As has already been remarked above,
there are references already in Rabbinic literature to the existence of Bible scholars
(baale ha-miqra), who are likely to the forerunners of the Masoretes.
All this suggests that Karaite scholars joined forces with an existing stream of
tradition of Bible scholarship in Rabbanite Judaism, enhancing it and developing it.
The evidence suggests that this took place mainly towards the end of the Masoretic
period in the second half of the 10th century and the 11th century. Their particular
contribution to the Tiberian Masoretic tradition was (i) to sponsor the safekeeping
of the model Masoretic codices produced by the Masoretes and, particularly in the
20

Kitb alAnwr (ed. Nemoy 1939), 554.

(1974, 8384) and Chiesa (1979, 58).

21 Cf. Bacher

22 See the document published by Mann

1970, 2:4344 .

11

11th after the cessation of the activities of the Tiberian Masoretes, the production of
accurate copies of Masoretic bibles; (ii) the production of some Masoretic treatises
and (iii) the development of the para-masoretic philological activity of grammar.
As we have seen, the inscriptions on several of the Bible codices that were
dedicated to Karaite communities indicate that they were to be used as models for
scribes to consult. The inscription on the Aleppo codex indicates that the
manuscript should be made available to both Karaites and Rabbanites for this
purpose. Several of the inscriptions also indicate that the codices should be used for
liturgical reading by the Karaite communities on sabbaths and festivals. This is
found, for example, in the Aleppo codex (
in order that they bring it
[the codex] out to the settlements and communities in the holy city on the three
pilgrimage festivals, the festival of Passover, the festival of Weeks and the festival
Tabernacles to read in it (Kahle 1927, 4)), the Cairo Codex of the Prophets (

This is the codex, the Eight Prophets, which

Yabe ben Shlomo consecrated in Jerusalem for the Karaites who celebrate the
feasts at seeing the moon, for them all to read on Sabbbath days, at new moons and
at the feasts. (Kahle 1947, 11214),23 Cod. 34 of the II Firkovitch collection (

This bible should be taken to one of the settlements in which

there are Karaite communities on sabbaths and festivals in the city of Cairo so that
the congregation can read it each sabbath and blessed festival (Kahle 1927 no. 13,
pp. 74-77)). The use of masoretic codices for liturgical reading distinguished the
Karaites from the Rabbanites, who continued to use scrolls for this purpose (1979).
The introduction of the codex in the Islamic period for the writing of Jewish
scripture is likely to have been influenced by the use of the codex for the writing of
the Qurn. This is reflected in the medieval Hebrew term for codex, viz. , which
is clearly a loan of the Arabic term muaf (Khan 2013, 67). The use of the codex by
the Karaites for liturgical reading can be interpreted as a reflection of a further
rapprochement to the Islamic environment. The Rabbanites, by contrast, remained
more conservative and restricted the codex to non-liturgical use.

23 The Cairo Codex of the Prophets was preserved down to modern times in the Karaite synagogue in

Cairo. The consecration to the Karaites of Jerusalem should be dated to the 11th century. This manuscript
also has a colophon attributing it to the Masorete Moshe ben Asher, the father of Aharon ben Asher, with
the date 895 C.E.. There is now a consensus that the manuscript was written later, most likely in the 11th
century, and this a later copy of an earlier colophon. For the arguments regarding its dating, see Cohen
1982 , Glatzer 1989, 25059 , Lipschtz 1964, 67 .

12

An even greater degree of rapprochement of the Karaites with the Muslim

environment is reflected by an extant corpus of Hebrew Bible manuscripts that were
written by Karaites in the Middle Ages in Arabic transcription.
Most of the known manuscripts containing Karaite transcriptions of Hebrew
into Arabic script are found in the British Library (Hoerning 1889; Khan 1987, 2533;
1993), the Firkovitch collections of the National Library of Russia in St. Petersburg
(Harviainen 1993a; 1993b; 1994; 1995; 1996a), and in the Cairo Genizah collections
(Khan 1990).24 These manuscripts emanate from Palestinian circles of Karaites or
Karaites in Egypt who had migrated to Egypt from Palestine after the capture of
Jerusalem by the Crusaders in 1099. The majority of them were written in the 10th
and 11th centuries. One of the transcriptions in the British Library (Or. 2554) has a
colophon which states that it was written in Ramle in 395 A.H. (1004-5 C.E.). Several
of the other manuscripts of the British Library corpus are written with the same
form of script and orthography. These include Or. 2548. Or. 2550, Or. 2551 fol 31-56,
Or. 2551 fols. 57-101, Or. 2581A fols. 31-46. It would appear that these manuscripts all
come from the pen of the same scribe. The manuscripts were preserved down to
modern times mainly in Egypt. This applies, of course, to the material from the
Cairo Genizah. The British Library corpus comes from a collection of mainly Karaite
manuscripts that were purchased by the library from the bookseller M. W. Shapira
of Jerusalem in 1882. The main source of Shapiras manuscripts and also the
manuscript acquired by Abraham Firkovitch appears to have been the Karaite
community of Cairo. A number of transcription fragments preserved in the Cairo
Genizah were originally parts of manuscripts acquired by Shapira and now in the
British Library, which shows that these British Library manuscripts must have come
from Cairo (Khan 1990, 34). Some of the transcriptions in the British Library and
Firkovitch collection may have been acquired also from the Karaite community of
the Iraqi town of Ht on the Euphrates (Hoerning 1889, v; Harviainen 1991; 1996b;
1998)
The transcriptions represent a convergence with the external form of the
Muslim Arabic Quran and also with the concepts of authority associated with the
transmission of Muslim scripture. Unlike the Masoretic Bible codices, which were
used by both Rabbanites and Karaites, these transcriptions were exclusive to the
Karaites.
The authoritative written form of Muslim scripture was fixed in the early
Islamic period. This was known as the Uthmanic text, since it was based on a codex
(muaf) authorized by the caliph Uthman in the first century A.H./7th century C.E.
(1905, 647) mentions the existence of an Arabic transcription of part of the book of Daniel in
the Khedevial Library of Cairo now the Egyptian Library .

24 Gottheil

13

To be precise, what was authorized was the rasm of the Uthmanic text, i.e. the shape
of the letters, though not the diacritical points. In early Quran manuscripts the
diacritical points of the Arabic letters are, in fact, frequently omitted. Although the
rasm became fixed, this could potentially be read in various ways and a variety of
reading traditions (qirt) existed. The qira was regarded as the authoritative core
of the text of scripture, which was based on the matrix of the rasm. It was crucially
important, therefore, to establish principles for determining which qirt were
authoritative.
The early generations of Quran readers felt a considerable amount of freedom
in determining the reading of the Uthmanic fixed consonantal text. They often
adopted one reading of the consonantal text rather than another on the basis of
their judgment of its grammatical correctness, unconstrained by any other criteria
(No ldeke, Bergstrasser, and Pretzl 1938, 120; Beck 1946, 188). By the time of the
Abbasid period, however, in the middle of the second century A.H., the freedom
allowable in the choice of Quranic readings began to be narrowed down. This was
achieved by the introduction of two conditions for the selection of a reading, in
addition to the requirement that it be grammatical and in conformity with the fixed
consonantal text: the condition that the reading must be based on the normative
usage of prestigious readers of earlier times, and that the reading must be agreed
upon by a majority of readers. The two conditions were not necessarily mutually
exclusive; they were both aspects of the concept of a generally agreed practice. The
sources of authority for establishing the correct reading of the consonantal text of
the Quran which are recognized by Sbawayh (d. 180 A.H./796797 C.E.) are those of
the majority (qirat al-mma) and the model of former authoritative sources (alsunna). He, in fact, identifies the one with the other, as is shown by his statement:
al-qira l tulaf liannah al-sunna The reading [of the majority] is not to be
disputed, because it is the normative usage.25 This expresses the view that the
majority reading has religious sanction, since it is the normative ideal usage of the
community. This notion of sunna and its merging with consensus is found also in
the doctrine of the ancient schools of Islamic jurisprudence before al- Shfi (d. 204
A.H./820 C.E.) (Schacht 1950, 5881). Al-Farr (d. 207 A.H./822 C.E.) also regards the
agreement of the majority of the readers and the traditions of the ancients as
sources of authority for establishing the correct reading. When referring to these, he
generally uses the terms ijtim and r.
Throughout the third century A.H., the so-called majority principle was
widely used to establish the authoritative qira of the Quranic text. This was due
mainly to the work of Ab Ubayd (d. 224/834) and Ab H tim (Sahl ibn
25

Kitb, ed. H. Derenbourg Paris, 1881 , 1:6222.

14

Muammad) al-Sijistan (d. 255/ 869). The application of the majority principle in
the selection of readings excluded those of small minorities. In cases where there
was no agreement by a clear majority, Ab Ubayd, Ab H tim and others
restricted their notion of majority to that of the readers of specific centres, such as
Medina and Kfa, or Medina and Mecca, or to that of specific readers, such as Nfi
and A im.
By the fourth century A.H., under the instigation of Ibn Mujhid (d. 324/936),
the tradition principle, whereby authority was given to the tradition of specific
readers, began to replace the majority principle. Ibn Mujhid established seven
canonical traditions of reading, which were endorsed by the ruling Abbsid
regime.26 These still fulfilled the requirements that they should conform to the rasm
of the authoritative text, that they should be grammatically correct, and that they
should be broadly authenticated. At a later period the seven canonical traditions
came to be accepted on the basis of their authoritative pedigree alone, in the
manner of the principles of establishing the authority of adth (traditions of the
sayings of the prophet Muammad). Ibn Mujhid himself applied some degree of
critical assessment of the content of the traditions, notably in their degree of
grammatical correctness (Nasser 2015).
We find eloquent evidence for the Karaites convergence with Muslim thought
regarding the transmission of scripture in the writings of the Karaite author
al-Qirqisn (first half of the tenth century C.E.). In a number of passages in his work
Kitb al-Anwr w-al-Marqib, he expresses his opinion about the basis of authority
of Hebrew scripture. He makes it clear that the authority lies in the text represented
by the reading tradition (qere) and not in that represented by the written tradition
(ketiv). Moreover the reading tradition derives its authority from the agreement of
the entire community (ijm) and not from the authority of the sages or that of any
specific group of people. The ketiv of the Hebrew Bible can be read in several
different ways and the correctness of one reading rather than another can only be
established by ijm. Furthermore in many cases reading the text on the basis of the
written tradition blatantly results in the wrong meaning (ed. Nemoy 1939, II.23.6).
For example, the frequent word in the Pentateuch would be read as boy rather
( Psa. 119.166)
than girl if the ketiv is followed. The word in
would have to be read with a in as if the ketiv is the basis of the reading and
the meaning would be I have broken rather than I hope.27
Al-Qirqisn was aware of the fact that there were some differences in reading
between the communities of Palestine and Iraq (ahl al-Shm wa-ahl al-Irq). In
Kitb alSaba f alQirt ed. Cairo, 1972 .
ketiv of the letter was regarded as being // and its reading as /s/was
considered to be the qere of this ketiv, i.e. the letter sin did not exist in the qere Steiner 1996 .
26 Ibn Mujhid,

27 In Rabbinic tradition the

15

such cases the reading of the community of Palestine must have the supreme
authority, even though the community of Iraq was larger. By the term ahl al-Shm
al-Qirqisn was referring to the Tiberian tradition of reading.
The position of al-Qirqisan with regard to the biblical text, therefore, is as
follows. The authoritative text of the Bible was represented by the reading tradition,
which was validated by the ijm of the entire nation (al-umma) in most of its
details. Where there was no overriding consensus in the nation as a whole with
regard to certain aspects of the tradition, it is the reading (qira) of the Palestinians
(ahl al-shm) that was the correct and authoritative one. That is to say, the correct
tradition in all its details is established by the ijm of the ahl al-shm rather than
that of the nation as a whole. This has clear parallels to the overriding authority
attributed to orally transmitted reading traditions (qirt) of the Quran and also to
the notion that ijm was a key determinant of the authority of a reading tradition.
Al-Qirqisans advocacy of ijm as a source of authority may have been further
reinforced by the influence of by Mutazil thought, which had a major impact on
medieval Karaite thought in general at this period.28 The Mutazila rejected tradition
as a source of law but accepted the validity of ijm.29 The Mutazil theologian Abd
al-Jabbr (320/932 to 414-16/ 1023-25) considered traditionalism (taqld) to be an
unsatisfactory way of acquiring knowledge, since it involved the uncritical
acceptance of a report without demanding proof or evidence.30 He maintained that
ijm, on the other hand, had probative value (ujjiyya). The probative value
followed from the existence of ijm. It does not require any proof that the
information it conveys is true.31
The adoption of the reading tradition as the overriding basis of authority had
the consequence that the Hebrew Bible could not be considered to offer two sources
of authority, one on the basis of the way it is read and the other on the basis of the
way it is written. The interpretation of the Scripture on two levels, one according to
28 It is well known that the Karaites were influenced by many doctrines of Muslim Mutazil theologians.
For Mutaliz ideas elsewhere in alQirqisns Kitb alAnwr, cf. BenShammai 1984, 27ff. .
29 alShfi, Kitb alUmm Bulaq, 132125/190307 , 7:252253; cf. Schacht 1950, 41, 25859 . Also
Ibn alMurta, Kitb abaqt almutazila, ed. S. DiwaldWilzer Beirut, 1961 , pp. 819, 9524; Abd al
Jabbr, Kitb fal al itizl waabaqt almutazila, ed. Fud Sayyid Tunis, 1393/ 1974 , p. 186.
30 AlMun f abwb altawd waladl Cairo, 138089/ 196069 , 12:123126; ar alul alamsa,
ed. Abd alKarm Umn Cairo, 1384/1965 , p. 61.
31 Mun, 17:199 fa'amm istidll al iat alijm min jihat alaql fabad As for the demonstration
of the validity of ijm by reason, this is unnecessary. . Cf. Bernand 1969; 1972 . A similar fideistic
acceptance of the probative validity of ijm and the rejection of traditions is expressed by Ab al
Husayn alBar, who was the pupil of Abd alJabbr, cf. Kitb almutamad f ul alfiqh Damascus,
1384/1964 , pp. 457540. Elsewhere Abd alJabbr states that ijm is supported by the Qurn and
sunna, cf. ar, p. 89. The extreme rationalist Mutazal alNaz zm and his school, however, had
misgivings about the reliability of ijm, on the grounds that information has to be supported and
ascertained before it can form the basis of ijm, i.e. ijm is only the consequence of truth, not the source
of truth, cf. Abd alQhir ibn T hir alBadd, Kitb ul aldn Istanbul, 1928/1346 , pp. 1920.

16

the ketiv and one according to the qere was a practice that is found in Rabbinic
sources. As shown by Naeh (1992; 1993), this was a phenomenon that developed in
the Talmudic period. It is reflected by the Talmudic dictum
The reading has authority and the traditional text has authority. The details
of the spelling of the written text, in particular the distribution of full and defective
orthography, were used as a source for interpretation in various Rabbinic texts
(Goldberg 1990). According to the Midrash Genesis Rabbah, for example, there is
exegetical significance as to why the second instance of the name Efron is spelled
without a waw in the verse Gen. 23.16 whereas the first instance of the name in the
verse and elsewhere in Gen. 23 has a waw:

Abraham agreed with Ephron; and Abraham weighed out for Ephron the silver
which he had named in the hearing of the Hittites, four hundred shekels of
silver, according to the weights current among the merchants. (Gen. 23.16).
The lack of waw (ketiv aser) indicates that Efron will suffer want because he
was envious and mean in accordance with the verse:

A miserly man hastens after wealth, and does not know that want will come
upon him. (Prov. 28.22)
In some cases such Midrashic texts exhibit a terminology and style of
presentation that constitute embryonic Masoretic notes regarding differences in
orthography of similar words (Martn Contreras 1999; 2002; 2003). As has been
remarked above, exegetical comments based on differences in orthography are
indeed found embedded within the Masoretic notes in some of the Tiberian codices,
e.g. the comment on the orthography of
( Job 8:8) in the Masorah of the Aleppo
codex discussed by Zer (2003).
If the qere is the only source of authority, as is the opinion of al-Qirqisn, then
variations in orthography cannot be a legitimate source of authoritative exegesis, as
in the comment on Job 8:8 in the Aleppo codex. Furthermore, convergence with the
Islamic model of scriptural authority would logically have resulted in the
inconsistency between the ketiv and the qere being considered problematic. One of
the key requirements of authoritative Quranic reading traditions at that period was
that they conform to the rasm of the written text. This was clearly not the case in
Hebrew scripture, in which the difference between ketiv and qere is sometimes very
substantial, including reading whole words that are not written and writing whole
words that are not read. The Hebrew Bible required a written rasm that
corresponded to the reading tradition. The Karaite transcriptions of Hebrew Bibles

17

represent the resolution of this tension by the abandonment of the traditional

Hebrew written text and the provision of an acceptable rasm. The new rasm was
created on the model of Islamic scripture in Arabic script. In addition to the use of
Arabic script the manuscripts of the transcribed Bible exhibit a convergence with
the Arabic Quran in codicological features, such as the use of red ink to mark
vocalization, the occasional use of Arabic vocalization signs and the insertion of
ornamentation at various points on the page that resemble what is found in
contemporary Quran manuscripts. Moreover some manuscripts even use the
Arabic word to represent the reading of the Tetragrammaton.
A few of the extant Karaite Bibles in Arabic script exhibit what is essentially a
letter for letter transliteration of the Hebrew orthography rather than a phonetic
transcription in Arabic orthography. It is likely that these are among the oldest
manuscripts. The Hebrew matres lectionis, for example, are reproduced in Arabic
script where they would not be appropriate according to Arabic orthography, e.g.

( BL Or 2541, fol. 18v, 11 || L Ex. 13.1 saying).
( BL Or 2541, fol. 30v, 13 || L Ex. 17.9 top of).
( BL Or 2541, fol. 16v, 9 || L Ex. 12.29 the night).

Conversely Arabic matres lectionis are not used when they are lacking in the
Hebrew text even where they would be required in Arabic orthography to represent
long vowels, e.g.
( BL Or 2541, fol. 17r, 9 || L Ex. 12.33 the people).
( BL Or 2541, fol. 17r, 2 || L
Ex. 12.36 favour).

It is important to note, however, that such texts represent the qere in places
where there is a conflict of qere and ketiv, e.g.
( BL Or 2541, fol. 26v, 4 || L Ex. 16.2 and they (mpl.)

murmured).
( BL Or 2541, fol. 29r, 8 || L Ex. 16.7 you (mpl.) will

murmur).
Moreover, there is a tendency to eliminate the inconsistency that is found in
the distribution of the matres lectionis waw and yod of the Hebrew ketiv, to which
exegetical significance was attached in Rabbinic sources. This is seen in the fact that
in many contexts where waw and yod are inconsistently used in the Hebrew ketiv,
the Arabic text more regularly uses the corresponding Arabic mater lectionis, ww
and y. This results in the Arabic matres lectionis being used in many cases where
the Hebrew ketiv has defective orthography, e.g.
( BL Or 2541, fol. 17r, 2 || L Ex. 12.30 great (fsg)).

18

( BL Or 2541, fol. 17r, 12 || L

Ex. 12.34 in their garments).

( BL Or 2541, fol. 21v, 5 || L Ex. 14.7 officers).
( BL Or 2541, fol. 26r, 13 || L Ex. 16.1 from Elim (nl)).

These early Karaite Bibles, therefore, attempted to produce a rasm that

corresponded to the reading tradition and had an internal consistency.
The majority of extant Karaite Bibles in Arabic script, however, use a
transcription system that is based essentially on the orthographic practices of
Classical Arabic. These use matres lectionis to represent all long vowels, as is the
practice in Arabic orthography. This appears to be a later development. Indeed a
few manuscripts exhibit a hybrid system of orthography, which include features
both of Hebrew orthography and Arabic orthography. These represent a transitional
stage of development between the Hebrew type of orthography and the Arabic type
of orthography, e.g.
( BL Or 2551 fol. 34v, 13 || L Deut 21.7 wife).
( BL Or 2539 MS B, fol. 115r, 1 || L Num. 6:18 head)

In general the written transcription in the Karaite Bibles could not be read
correctly without a knowledge of the reading tradition. This is shown by the fact
that they often omit crucial details, such as diacritical points on the Arabic
consonants. Moreover the spelling is often ambiguous. Each Arabic mater lectionis,
for example, represents different qualities of Hebrew vowel.
In principle the traditional ketiv of Hebrew scripture could have been adjusted
to conform to the qere without changing the script. Indeed this is found in some
fragments of Bible manuscripts from this period that were written for private use
and have been preserved in the Cairo Genizah.32 The adoption of the Arabic script in
the transcriptions reflects a convergence with the external form of the Quran,
which was facilitated by the assignment of exclusive authority to the reading
tradition.
The Islamic model required not only conformity of the reading to the rasm of
the written text but also conformity of the reading to Arabic grammar. It was
important for the Karaites, therefore, to legitimate the grammatical integrity of the
Tiberian reading tradition. As has been remarked above, in the 10th century a
tradition of Karaite Hebrew grammatical thought emerged. The main source of our
knowledge of this Karaite grammatical tradition in its early stages of development in
the 10th century is the grammatical commentary on the Bible by Ysuf ibn N
known as the Diqduq (Khan 1998; 2000). Ibn N does not offer a systematic
description of Hebrew grammar but rather concentrates on points that he believed
32 GoshenGottstein

1962, 39ff. , DezMacho 1971, 92 , Yeivin 1980, 3031 .

19

may be problematic for the reader or concerning which there was controversy. One
of the central concerns was to show that apparent inconsistencies in similar words
can be explained as conforming to rational rules of grammar. This was often
achieved by arguing that similar forms differing in small details were derived from
different morphological bases. In the class of verbs which we refer to as final
geminates, for example, there is variation in the position of stress in the past forms,
e.g. qa ll they are swift (Job 9.25) vs. w-qallu and they are swift (Hab. 1.8).
According to Ibn N this is not an arbitrary variation, but rather the forms with the
penultimate stress are derived from a noun base whereas the forms with final stress
have an imperative base. The Karaite grammarians were concerned only with the
Tiberian reading tradition and did not take into account of the ketiv. Their work
vouchsafed the grammatical integrity of the reading tradition.
The Arabic transcriptions of the Hebrew Bible represent an extreme case of
convergence with the Islamic environment. As has been remarked, there was
variation in the Karaite community as to the degree of convergence. During the
period in which the transcriptions were made, Karaites used also Hebrew script for
both Arabic and Biblical Hebrew. Likewise there is evidence that some Karaites
maintained the Talmudic principle of . The Karaite
lexicographer David ben Abraham al-Fs (second half of the 10th century), for
example, cites this as a legitimate principle of exegesis in the introduction to his
lexicon Kitb Jmi al-Alfz (ed. Skoss, I, 1-13).33
Of course convergence with the culture of the Islamic environment was not
unique to the Karaites. Rabbanites scholars of the 10th and 11th centuries adopted the
Arabic language in their writings and many elements of the Islamic intellectual
tradition. This included a preference for the qere of the Hebrew Biblical. Saadya
Gaon, for example, in principle follows the qere in his Arabic translations of the
Bible and exegesis. The extent of the convergence, however, was to a lesser degree
and this is reflected by the fact that they maintained Hebrew script in their writings.

References
Allony, Nehemiah. 1964. Reimat Munax im Qarait Me-Ha-Mea Ha-eminit. In
Kitve Ha-X evra Le-X eqer Ha-Miqra Be-Yisrael Le-Zexer Dr Y.P. Korngreen,
edited by Asher Weiser and Ben-Zion Luria, 32463. Tel-Aviv: Niv.
. 1979. . Beit Mikra
78: 32134.
33 It was also accepted by the Byzantine Karaite scholar Judah Hadassi

113 .

12th century ; cf. Bacher 1895,

20

Bacher, Wilhelm. 1895. Jehuda Hadassis Hermeneutik Und Grammatik.

Monatsschrift Fr Geschichte Und Wissenschaft Des Judentums 40: 10926.
. 1974. Die Anfnge Der Hebrischen Grammatik; and Die Hebrische
Sprachwissenschaft Vom 10. Bis Zum 16. Jahrhundert. Amsterdam Studies in
the Theory and History of Linguistic Science, v. 4. Amsterdam: J. Benjamins.
Beck, Edmund. 1946. Arabiyya, Sunna Und mma in Der Koranlesung Des Zweiten
Jahrhunderts. Orientalia 15: 180224.
Ben-Shammai, Haggai. 1984. The Attitude of Some Early Karaites towards Islam. In
Studies in Medieval Jewish History and Literature, edited by Isadore Twersky,
2:340. Cambridge, MA: Harvard University Press.
Ben-Zvi, Izhak. 1960. The Codex of Ben Asher. Textus 1: 116.
Bernand, Marie. 1969. Lim Chez Abd Al-abbr et Lobjection Dan-Nam.
Studia Islamica 30: 2738.
. 1972. Nouvelles Remarques Sur Lim Chez Le Q Abd Al-abbr.
Arabica 19: 7885.
Chiesa, Bruno. 1979. The Emergence of Hebrew Biblical Pointing: The Indirect Sources.
Judentum Und Umwelt 1. Frankfurt: Lang.
Cohen, Menahem. 1982. Has the Cairo Codex of the Prophets Indeed Been Written
by Moshe B. Asher? Alei Sefer 10: 512 (in Hebrew).
Dez Macho, Alejandro. 1971. Manuscritos hebreos y arameos de la Biblia:
Contribucin al estudio de las diversas tradiciones del tecto del Antiguo
Testamento. Rome: Institutum patristicum Augustinianum.
Dotan, Aron. 1997. Dawn of Hebrew Linguistics: The Book of Elegance of the Language
of the Hebrews. Jerusalem: ha-Iggud ha-Olami le-Madae ha-Yahadut.
Dotan, Aron, 1928-. 1977. Ben Ashers Creed: A Study of the History of the Controversy.
Vol. 3. Masoretic Studies; Missoula, Mont.: Published by Scholars Press for the
Society of Biblical Literature and the International Organization for
Masoretic Studies,.
Eldar, Ilan. 1980. On Ben-Asher and Ben-Naftali. Lonnu 45: 311313 (in Hebrew).
. 1994. The Study of the Art of Correct Reading as Reflected in the Medieval
Treatise Hidyat Al-Qri. Jerusalem: Academy of the Hebrew Language (in
Hebrew).
Erder, Yoram. 1994. Early Karaite Conceptions about Commandments Given before
the Revelation of the Torah. Proceedings of the American Academy for Jewish
Research 60: 10140.
Glatzer, Mordechai. 1989. The Aleppo Codex Codicological and Paleographical
Aspects. Sefonot 4: 167276 (in Hebrew).

21

Goldberg, Arnold. 1990. The Rabbinic View of Scripture. In A Tribute to Geza

Vermes. Essays on Jewish and Christian Literature and History, edited by Philip
R. Davies and Richard T. White, 15366. Sheffield: JSOT Press.
Goshen-Gottstein, Moshe. 1960. The Authenticity of the Aleppo Codex. Textus 1: 17
58.
. 1962. Biblical Manuscripts in the United States. Textus 2: 2859.
Gottheil, Richard. 1905. Some Hebrew Manuscripts in Cairo. Jewish Quarterly
Review 17: 60955.
Graetz, Heinrich. 1881. Die Anfnge Der Vocalzeichen Im Hebrischen.
Monatsschrift Fr Geschichte Und Wissenschaft Des Judentums 30: 34867.
Hadassi, Judan ben Elijah ha-Abel. 1836. Sefer Eshkol Ha-Kofer. Eupatoriy: Mordecai
Tirisken.
Harviainen, Tapani. 1991. Abraham Firkovitsh, Karaites in Ht, and the Provenance
of Karaite Transcriptions of Biblical Hebrew Texts in Arabic Script. Folia
Orientalia 28: 17991.
. 1993a. Karaite Arabic Transcriptions of Hebrew in the Saltykov-Shchedrin
Public Library in St. Petersburg. In Estudios Masorericos. En Memoria de
Harry M. Orlinsky., edited by Emilia Fernandez Tejero and Mara Teresa
Ortega Monasterio, 6372. Xtos Y Estudios Cardenal Cisneros de La Biblia
Polglota Matritense (TECC 55). Madrid: Instituto de Filologa del CSIC.
. 1993b. Karaite Bible Transcription with Indiscriminate Use of Tiberian
Pata and Segol Vowel Signs. In Semitica. Serta Philologica Constantino
Tsereteli Dicata, edited by Riccardo Contini, Fabrizio Pennacchietti, and
Mauro Tosco, 8397. Pubblicazioni Del Gruppo Di Ricerca Lessicografia
Semitica E Lessico Ebraico Finanziato Dal C.N.R. 6. Torino: Silvio Zamorani
Editore.
. 1994. A Karaite Bible Transcription with Indiscriminate Counterparts of
Tiberian Qame and olam (Ms. Firkovitsh II, Arab.-evr.1). In Proceedings of
the Eleventh Congress of the International Organization for Masoretic Studies
(IOMS), Jerusalem June 21-22, 1993, edited by Aron Dotan, 3340. Jerusalem:
The World Union of Jewish Studies.
. 1995. A Karaite Letter-for-Letter Transliteration of Biblical Hebrew - MS
Firkovitsh II, Arab.-Evr. 355. Textus 18: 196177.
. 1996a. MS Arab.-Evr. 2 of the Second Firkovitsh Collection, A Karaite Bible
Transcription in Arabic Script. In Studies in Hebrew and Jewish Languages
Presented to Shelomo Morag, edited by Moshe Bar-Asher, *41-*59. Jerusalem:
Bialik Institute.
. 1996b. The Cairo Genizot and Other Sources of the Firkovich Collection in
St. Petersburg. In Proceedings of the Twelfth International Congress of the

22

International Organization for Masoretic Studies, edited by E. John Revell, 25

36. Atlanta, GA: Scholars Press.
. 1998. Abraham Firkovich and the Karaite Community in Jerusalem in 1864.
Manuscripta Orientalia 4 (2): 6670.
Hoerning, Reinhart. 1889. British Museum Karaite MSS. Descriptions and Collation of
Six Karaite Manuscripts of Portions of the Hebrew Bible in Arabic Characters;
with a Complete Reproduction ... of One, Exodus I. 1 - VIII. 5, in ... Facsimiles.
London.
Ibn Mujhid, Amad ibn Ms. 1972. Kitb Al-Saba F Al-Qirt. Edited by Shawqi
ayf. Cairo: Dr al-Marif.
Kahle, Paul. 1927. Masoreten Des Westens. 2 vols. Texte Und Untersuchungen Zur
Vormasoretischen Grammatik Des Hebrischen. Stuttgart: W. Kohlhammer.
. 1947. The Cairo Geniza. Schweich Lectures of the British Academy. London:
Cumberlege [for the British Academy],.
Khan, Geoffrey. 1987. Vowel Length and Syllable Structure in the Tiberian Tradition
of Biblical Hebrew. Journal of Semitic Studies 32 (1): 2382.
. 1990. Karaite Bible Manuscripts from the Cairo Genizah. Cambridge
University Library Genizah Series. Cambridge: Cambridge University Press.
. 1993. The Orthography of Karaite Hebrew Bible Manuscripts in Arabic
Transcription. Journal of Semitic Studies 38 (1): 4970.
. 1998. The Book of Hebrew Grammar by the Karaite Joseph Ben Noa.
Journal of Semitic Studies 43 (2): 26586.
. 2000. The Early Karaite Tradition of Hebrew Grammatical Thought:
Including a Critical Edition, Translation and Analysis of the Diqduq of Ab
Yaqb Ysuf Ibn N on the Hagiographa. Studies in Semitic Languages and
Linguistics. Leiden: Brill.
. 2003. The Contribution of the Karaites to the Study of the Hebrew
Language. In A Guide to Karaite Studies: The History and Literary Sources of
Medieval and Modern Karaite Judaism, edited by Meira Polliack, 291318.
Handbuch Der Orientalistik 73. Boston: Brill.
. 2013. A Short Introduction to the Tiberian Masoretic Bible and Its Reading
Tradition. 2nd ed. Piscataway, NJ: Gorgias.
. 2014a. The Medieval Karaite Tradition of Hebrew Grammar. In A Universal
Art. Hebrew Grammar Across Disciplines and Faiths, edited by Vidro, Nadia,
Irene E. Zwiep, and Judith Olszowy-Schlanger, 1533. Leiden: Brill.
. 2014b. The Relationship of Hidayat Al-Qari to the Karaite Grammatical
Tradition. In Studies in Hebrew and Related Fields Presented to Ilan Eldar,
edited by Moshe Bar-Asher and Irit Meir, 277283 (in Hebrew). Jerusalem:
Karmel.

23

Khan, Geoffrey, Maria A ngeles Gallego, and Judith Olszowy-Schlanger. 2003. The
Karaite Tradition of Hebrew Grammatical Thought in Its Classical Form: A
Critical Edition and English Translation of Al-Kitb Al-Kf F Al-Lua AlIbrniyya by Ab Al-Faraj Hrn Ibn Al-Faraj. Studies in Semitic Languages
and Linguistics 37. Leiden: Brill.
Klar, Benjamin. 1954. . Edited by Abraham Meir
Habermann. Tel-Aviv: Mabarot le-Sifrut.
Lipschtz, Lazar. 1964. Kitb Al-Khilaf the Book of the illufim: Mishael Ben
Uzziels Treatise on the Differences between Ben Asher and Ben Napthali.
Textus 4: 229.
. 1965. Kitb Al-Khilaf: Mishael Ben Uzziels Treatise on the Differences between
Ben Asher and Ben Naphtali. Kitvey Mifal Ha-Miqra el Universiah Ha-Ivrit.
Jerusalem: Magnes.
Mann, Jacob. 1970. The Jews in Egypt and in Palestine under the Ftimid Caliphs. Vol.
2. New York: Ktav.
Martn Contreras, Elvira. 1999. Terminologa masortica en la exgesis de Gnesis
rabba (secciones Beresit y Noah). Sefarad 59 (2): 34352.
. 2002. Noticias masorticas en el midrs Lamentaciones Rabb . Sefarad
62 (1): 12541.
. 2003. Noticias masorticas en los midrasim halkicos ms antiguos y su
comparacin con los midrassim exegticos. Sefarad 63 (1): 11939.
Naeh, Shlomo. 1992. Did the Tannaim Interpret the Script of the Torah Differently
from the Authorized Reading? Tarbiz 61: 401448 (in Hebrew).
. 1993. En Em Lammasoret Second Time. Tarbiz 62: 455462 (in Hebrew).
Nasser, Shady. 2015. Revisiting Ibn Mujhids Position on the Seven Canonical
Readings: Ibn mirs Problematic Reading of Kun Fa-Yakna. Journal of
Quranic Studies 17 (1): 85113.
No ldeke, Theodor, Gotthelf Bergstrasser, and Otto Pretzl. 1938. Geschichte Des
Qorns. 2, part 3 ed. Leipzig: Dieterichsche Verlagsbuchhandlung.
Ofer, Yosef. 1989. M.D. Cassutos Notes on the Aleppo Codex. Sefunot 68: 277344.
Penkower, Jordan. 1981. Maimonides and the Aleppo Codex. Textus 9: 39128.
. 1989. A Tenth-Century Pentateuchal MS from Jerusalem (MS C3),
Corrected by Mishael Ben Uzziel. Tarbiz 58: 4974 (in Hebrew).
Pinsker, Simah. 1860. Lickute Kadmoniot. Zur Geschichte Des Karaismus Und Der
Karaschen Literatur. Wien.
Poznanski, Samuel. 1913. . Edited by Abraham Moses
Luncz. Jerusalem 10: 83116.
Qirqisani, Yaqu b ibn Ish aq. 1939. Kitab Al-Anwr W-Al-Marqib. Edited by Leon
Nemoy. New York: The Alexander Kohut Memorial Foundation.

24

Schacht, Joseph. 1950. The Origins of Muhammadan Jurisprudence. Oxford:

Clarendon Press.
Steiner, Richard. 1996. Ketiv-er or Polyphony: The - Distinction according to
the Masoretes, the Rabbis, Jerome, Qirqisn, and Hai Gaon. In Studies in
Hebrew and Jewish Languages Presented to Shelomo Morag, edited by Moshe
Bar-Asher, *151-*179. Jerusalem: Bialik Institute.
Vidro, Nadezda. 2009. Verbal Morphology in the Karaite Treatise on Hebrew
Grammar Kitb Al-Uqd F Tarf Al-Lua Al-Ibrniyya. Cambridge:
University of Cambridge.
Vidro, Nadia. 2011. Verbal Morphology in the Karaite Treatise on Hebrew Grammar:
Kitb Al-Uqd F Tarf Al-Lua Al-Ibrniyya. tudes Sur Le Judasme
Mdival 51. Leiden: Brill.
Yeivin, Israel. 1962. The Vocalization of Qere-Kethiv in A. Textus 2: 14649.
. 1980. Introduction to the Tiberian Masorah. Masoretic Studies. Missoula:
Scholars Press.
Zer, Rafael Isaac. 2003. Was the Masorete of the Aleppo Codex of Rabbinite or
Karaite Origin? Sefunot 8: 57387.
Zucker, Moshe. 1958. Against Whom Did Seadya Gaon Write the Polemical Poem
Essa Meshali? 27: 6182 (in Hebrew).