Hebrew Philological Practices from Antiquity to the Middle Ages

Geoffrey Khan, Hindy Najman, and Ishay Rosen-Zvi

The Forum provides a home for substantial original and creative works of scholarship with the goal of bringing profound research to the public and making specialist scholarship accessible to the broader academy.

A  PDF version is available with a complete list of references.  Authors’ Note: This essay originated in connection with a project on a lexicon of philological terminologies and benefited from feedback from the editors of that project (Anne Eusterschulte, Martin Kern, and Glenn Most).

It is almost a truism today that philology – in the sense of a systematic reflection on textuality and language – did not penetrate Hebrew culture before the Middle Ages, under new Islamic and Hellenistic influences. The traditional, oral and self-sufficient nature of the cultures of Hebrew textuality prevented them, so the conventional story goes, from developing philological theories and techniques similar to what we find in the surrounding cultures. This essay strives to question and amend this narrative. Through a review of Hebrew textual practices, both creative and interpretive, from the ancient periods to medieval times we offer a picture of a textual culture that exhibits both a tenacious spirit of survival and an accommodating tendency as it is forever interacting with surrounding cultures and traditions. Hebrew textual culture both preserves past traditions, and absorbs the practices of writing, scribalism, literary composition, and hermeneutics in new cultural and linguistic contexts. From antiquity until the present, Hebrew textuality is innovative and responsive, linguistically and philologically.


The chronological range of our essay is from the earliest manuscripts in the first millennium BCE (The earliest attested Hebrew inscriptions date from the first half of the first millennium BCE. However, our earliest Hebrew manuscripts are palaeographically dated to the third century BCE.) up until the manuscript sources of the Middle Ages. Extant sources across this long-ranging period attest to synchronically and diachronically interconnected practices of philological activity.

Our understanding of philology is a broad and variegated range of activities related to texts, their reading and production, codicological and scribal practices, as well as metatextual activities such as commentary, textual development, grammar, and translation. We understand “text” and philological activity as referring both to written forms of texts (e.g., Torah and Psalms) and oral forms (e.g., Mishnah, Talmud, Midrash, biblical reading traditions and liturgy). Although all of our extant sources were ultimately written down, we are nevertheless presenting a dynamic and interconnected oral dimension in our overarching definition of Hebrew philology, much of which was textualized in written form in the Middle Ages.

In this essay, we emphasize continuity and discontinuity from Antiquity to the Middle Ages. Throughout this period, Hebrew philological activity was influenced by the cultural environments in which it took place. In Antiquity, the predominant environment was Hellenistic and in the Middle Ages it was Islamic. In between these two periods we have a variety of cultural influences that are more local: Roman and then Byzantine Greek in the West and Sassanian Iranian in the East.

In order to understand the nature of these influences, it is essential that we consider in intimate and detailed ways the philological practices of the contemporaneous cultures as elsewhere represented across this lexicon project. While the practices are reshaped and informed by a variety of cultural contexts, there is nevertheless continuity in the written and oral traditions across the period that concerns us. We have various continuities of types of philological activity. Firstly, in the medieval period the oral traditions from the earlier periods, e.g., rabbinic literature and reading traditions of the Hebrew Bible, were textualized. For this reason, medieval sources give us a window onto Antiquity. Secondly, we can study in great depth the types of philological activities in the medieval period that are evident in abundance in textualized forms; these provide us with deeper insight into the nature of corresponding philological practices in the earlier period, which has less empirical data for primary sources. For these two reasons, we believe that there is a heuristic value in studying the whole chronological range, whereas the current state of scholarship fragments the history of Hebrew philology into its discrete cultural contexts and periods. Our vision is to join together these worlds which we believe are parts of a single organic development, albeit changing under the influence of and in conversation with the various cultural environments.

The extant sources exhibit a variety of types of activities that are associated with reading and understanding. By way of example, traditions of liturgy reach back to some of our earliest texts from the Hellenistic period, with respect to both performance and the form and function of prayer itself. By studying the Middle Ages we can understand much about Antiquity as well – through transmission and copying. Material and typological continuity are more discernable in the medieval period; this is relevant both to continuity of content (e.g., liturgy) and material continuity.

We will now turn to three distinctive stages of the development of Hebrew and corresponding philological practices as we reflect on how each of these exhibit the above phenomena. We will divide our essay into three formative periods. We refer to the activity of the rabbis in approximately 100-600 CE as the “intermediate period” in order to emphasis their role as a bridge between the textual activities of the ancient and the medieval periods.


Our earliest material evidence for philological and scribal activity in the Hellenistic period is most extensively from the Dead Sea Scrolls. There is some evidence already in the Assyrian, Babylonian, and Persian periods, but only in the form of inscriptions. The entirety of the Hebrew Bible is replete with discussions (sometimes implicit and at other times explicit) of scribes, writing, preservation, and the dissemination of scribal traditions. Across the biblical traditions there are a multitude of references to prophetic revelations, scribal dictations, royal records, and marriage documents. These are varied and extensive, such as e.g., gillayon, sefer, luaḥ, and megillah. The documents are at once human, divine, earthly, heavenly, buried, revealed, and they even fly. They are consumed, hidden, burnt, rewritten, expanded, and inscribed. Texts are themselves life-giving but can also kill. The gift of writing can be authorized and divine, but it can also be transgressive (e.g., heavenly tablets and transgressive writing in Jubilees).

Extra-biblical sources from the Hellenistic period are also very attentive to and focused on writtenness and preservation. There is a great deal of interest devoted to the authority of various traditions through heavenly scribes, divine dictation, and angelic revelation.

Hebrew and Aramaic literature (e.g., Jubilees and Enoch, among other texts) are replete with references to writing and heavenly scribes. The book of Jubilees also fills its 50 chapters with an obsessive habit of recording and protesting that all of the traditions that are being written in this book are already inscribed. Thus, philological activity and authorization practices through writing and reliable dictation become a preoccupation already in the Hellenistic period. The discourse about a Torah that is deliberately unwritten or oral will only explicitly emerge with later rabbinic texts starting already with the Tannaitic period.

The role of scribes is also essential to this discourse. Scribes are described as inspired and authoritative in the Jewish Hellenistic prophetic, legal, and wisdom texts. Moses is described as the divine amanuensis who records the law (Jubilees and Philo) and Enoch is the first scribe to record heavenly history (Jubilees). There is Baruch who, as Jeremiah’s scribe, records the whole of Jeremiah’s exchange with God and then is instructed to add more (Jeremiah 36). There is also Ezra, who is the paradigmatic scribe who reveals the law again and performs another Sinai through interpretation and translation (Nehemiah 8).

The Dead Sea Scrolls comprise over a thousand identified manuscripts from the Hellenistic periods. The scrolls disclose extensive philological activity. From copying to correcting, from preserving to expanding, from innovating to refining – the activities at the site of Khirbet Qumran were extensive. We also have evidence of intense writing activity through the material remains of the writing tables and the ink wells and the attentiveness to preserving the scrolls. The processes were expensive and time-consuming. Some scroll remains in caves 1 and 11 were preserved with great care and attention as they were wrapped in shrouds and then buried in jars in the caves. Other caves left fragmented materials behind, especially in the case of cave 4. The bands and the ties for the scrolls also betray great care and attention. Furthermore, there are signs of scribal markings and education amongst these recovered scrolls.

The discoveries of the Dead Sea Scrolls along with information from the Septuagint confirmed a radical pluriformity with respect to versional difference of biblical texts. In some books, e.g. Jeremiah, the variation is significant, whereas the Pentateuch is relatively fixed. Parts of all of the books from the current canon of the Hebrew Bible have been recovered with the exception of the book of Esther. In addition, the texts discovered among the Dead Sea Scrolls include additional psalms that were included in the Great Psalms Scroll from cave 11; Pesharim and other interpretative texts that explicitly refer to the biblical corpus and interpret those next, new scriptural compositions that were not included in the canon of the Hebrew Bible; parts of the deuterocanonical texts, e.g., Ben Sira; and additional liturgical and wisdom texts from caves 4 and 11 as well as legal texts that pertain specifically to the community which refers to itself as the yaḥad, or the ʿosey hattorah (adherents to the Law). The groups of texts bear witness to the tremendous growth with respect to writing and performance that are characteristic of this period.

The figures of the Maskil (lit. instructor) and the Teacher of Righteousness mentioned in several scrolls from Qumran also clearly had a pedagogical role. Both figures had a central role with respect to teaching, writing, precise interpretation and the dissemination of that teaching. There is also a complex relationship between orality and writtenness across the scrolls. The performance of liturgical, interpretative, and wisdom traditions plays a central role during this period of intense activity. Furthermore, there is a blurring of lines between text and interpretation. Moving forward from Daniel and Zechariah, where we can already see explicit practices of commentary and compositional expansion, we have the growth of a particular practice of Pesher at Qurman. In these Pesharim we begin to see explicit commentary practices which explain, update, actualize, and activate lemmata from various prophetic and wisdom texts.

The life of the various scribes that are mentioned in the biblical corpus came to have discourses tied to them. So, for example, Baruch will continue to have texts associated with his name long after the span of his tenure as Jeremiah’s scribe. He is said to have trained Ezra and ultimately will have an entire text ascribed to him along with a rich interpretative tradition. Ezra’s reputation as scribe in the Persian period will grow so much that the rabbis will come to say that “if Moses had not received the Torah, Ezra would have been worthy to do so” (Tosephta). Other figures such as Solomon, Enoch, Jeremiah, and Isaiah will continue to have new texts and traditions associated with them as scribal and inspired figures as their discourses continue to expand.

Overall, the Hellenistic period signals practices of composition that help us understand the development of biblical, interpretative, and liturgical traditions. Writing, reading, and preservation become increasingly central to the Hellenistic Jewish context where there is an ongoing expression of recovery and preservation alongside memory.

These practices can be said to be essential to the transition into the rabbinic period. Moreover, we can speak to a crystallized period of philological activity which is mirrored only in our second pillar of activity, namely in the Middle Ages. The philological practices that we can trace in the Hellenistic period participate in the work of scribal activity and the recording of liturgical and interpretative traditions which continue to preserve practices that are oral but only accessible to us in their written dimensions. Furthermore, practices of editing and interpretation are deeply engaged in the process of understanding and explanation. Philology then is inextricably linked to the work of hermeneutics and to that end we will need to trace lines of continuity that we can mark from the late Persian and Hellenistic periods through the Middle Ages.

It is fundamental to our project of understanding Hebrew philological practices that we are always in the business of reconstructing the past practices that are part of the first traceable signs we have of explicit commentary and performance through our later expressions and characterizations of grammar, institution formation, commentary, the fixing of set texts and the vital and dynamic preservation of those texts through oral performance, explanation, and the fixing of the text for the purposes of the synagogue and the study house.


Unlike most Judean groups in the Hellenistic and Roman periods, the rabbis kept their literary activity exclusively oral (this is probably the case also for the Pharisees, whom classical scholarship identified as the rabbis’ ideological precursors, and who disappeared from our sources after the destruction of the Temple in 70 CE). This literary activity was initially dedicated to expounding the law, based on inherited traditions. From the middle of the second century CE, however, this activity concentrated on the interpretation of the biblical texts (named כתבי הקדש, kitve haqodesh, “the holy books”), which they treated as a unified composition. The figure generally identified in the sources with this development is Rabbi Akiva.

The Pentateuch was divided into five physical books, which were copied independently for learning, but were bound together in one scroll for ritualistic use in the synagogue (named ספר תורה sefer torah, “the book of the Torah”). The rest of the Hebrew Bible was never bound in one volume, but nonetheless the rabbis treated it as a unified text. They refused to consider their holy writ as a collection of books, instead forming an image of a single book. The term torah in rabbinic literature, therefore, is used in the sense either of the Pentateuch or of the whole Bible, and may in some contexts even include rabbinic teachings. The rabbis maintained the concept of a single book in the study house in two complementary ways. On the one hand they treated all the books of their canon as a single unit, reflected through the intertextual practice of midrash (“homiletical exegesis”), while excluding from the purview of their activities all the other books, which were named ספרים חיצוניים sefarim hiṣoniyyim, “external books” (cf. the Christian name: Apocrypha, hidden books). On the other hand, they maintained their own legislation and traditions in an exclusively oral form.

Rabbinic literature can thus be considered the first Jewish culture to focus on a “canon,” not in terms of a sealed text with a fixed version, but in another sense: of one, indispensable book at the center of the curriculum. The traditional image of a council in Yavne at which the final canonization of the Bible was officially decreed is therefore misleading.

With regard to Torah scrolls, the rabbis manifest a philological sensitivity that anticipates the more systematic engagement in medieval philology. In rabbinic literature, however, this philological awareness is still sporadic and mixed with mythical claims. Two examples are the story about ספר העזרה sefer haʿazara (“the scroll of the [Temple] court”), an official copy of the Pentateuch allegedly deposited in the Temple that was the basis of the stability of the text, and תיקוני סופרים tiqquney soferim (“scribal emendations”), changes ascribed to the ancient scribes in order to explain readings that seemed to be corrupt. History, myth and philology are mixed together in the rabbinic portrayal of the transformation and stabilization of the holy text. The rabbis also developed laws for writing Torah scrolls for ritualistic use, although these were not codified in the Mishnah, but only in a later tractate from the Geonic period (c. the eighth century), named מסכת סופרים massekhet soferim, “a tractate on scribes.”

One of the oldest questions concerning the formation of rabbinic law is whether rabbinic legislation should be perceived as an actual product of biblical interpretation, as has been maintained since the advent of Wissenschaft des Judenthums (“the science of Judaism”) in the nineteen century, or whether it stems from an ancient oral tradition, which was only later superimposed upon Scripture. Following the seminal work of E. Urbach, many scholars tend to date Midrashic (i.e. homiletical) exegesis later than the study of free-standing halakhic (i.e. legal) tradition.

Ever since the young Abraham Geiger’s complaint in 1844 about the “opaque” exegetical sense of the rabbis, students of Midrash have labored to explain the essence of Midrashic hermeneutics and to reconcile it with contemporary perceptions of the nature of exegesis. Different explanations have been offered, including that the secret of Midrash, still known to the ancients, is a mystery to us; Midrash itself was given at Sinai; Midrash reveals the depth of the plain meaning; and various approaches that consider the practice of Midrash to have been formed ex post facto to buttress existing ideas and narratives. In order to read Midrash as exegesis, the very concept of “biblical exegesis” must be revised. Although the term פשוטו של מקרא (peshuṭo shel miqra) appears in rabbinic literature, it probably means there simply “the flow of the text” and does not refer yet to an exegetical method, as it does in the writings of medieval commentators, in Spain and France, from the tenth century on, where the term פשט peshaṭ is used with the sense of “simple or literal meaning.” Scholars have struggled to explicate Midrashic hermeneutics, and neologisms like “creative philology” (Isaac Heinemann) have been offered. Subsequent scholars offered more nuanced definitions, but their goal was similar: namely to account for the midrashists’ meticulous attention to every word in the Bible, together with intensified sensitivity to linguistic phenomena, which refuses, however, to distinguish between various layers of meanings, and instead, much unlike Hellenistic allegory, sees them all as existing on the same plain.

Although there is a continuation rather than a break between rabbinic exegesis and that of their predecessors, there are several noteworthy innovations.

(a) Unlike the blurring of text and interpretation, which was common in the Second Temple period, early rabbinic midrash distinguishes explicitly between the two, a distinction expressed by the Midrashic structure of a lemma followed by a commentary. A new consciousness is thus crystalized: a biblical text separated from its interpretation. We see here a development from Pesher, as a more explicit commentary tradition develops.

(b) In rabbinic Midrash we encounter new interpretive techniques, more developed and detailed than anything existing in Second Temple literature, in either Hebrew (e.g., Jubilees, the Dead Sea Scrolls) or Greek (Philo, Paul).

(c) These techniques are accompanied by a new terminology, marking interpretive strategies, such as אין לי אלא… מניין, “I only know [from the verse] X, whence do I learn also Y?,” after which an exegetical strategy to include also Y is offered. Note, however, that while Y is added exegetically, it is explicitly distinguished from X, as only the former is marked as explicit in the verse. This exegetical technique thus makes interpretation self-reflective. Furthermore, some of the formulaic strategies present alternatives, disparities, and disagreements (such as אתה אומר… או אינו אלא…, “you say [the correct interpretation is] X, or rather is it nothing else than Y?,” which demands explicating why an interpretive choice is better than the alternatives), thus demonstrating to the learners not just the conclusion of the exegetical endeavor, but also the process of interpretation itself and the decision-making that it involved. This method stands in stark contrast to the divinely inspired exegesis in the pesharim attested in the Dead Sea Scrolls, which allows no disagreement or exegetical alternatives.

(d) The early rabbis were also the first to systematically distinguish between the legal and narrative parts of the Pentateuch, and accordingly between two types of intellectual focus in the study house: הלכה (halakha, law) and אגדה (aggada, non-legal, moral and religious occupation). In the later rabbinic period this distinction developed into the creation of aggadic midrashim (“homiletical exegesis”) for the various parts of the Pentateuch. The earliest of these is Genesis Rabba, which was redacted in the fifth century CE.

(e) Maybe the most crucial innovation of the rabbis is the clear distinction between Mishnah and Midrash. While in pre-rabbinic literatures we find all sorts of combinations of exegetical activity and independent legal traditions (as, for example, in the genre named by scholars, not unproblematically: “rewritten Bible”), the rabbis clearly distinguished between two kinds of occupations in the study house. These were the independent juristic development of the law, codified in the Mishnah, and the systematic engagement with the Bible, in both its legal and narrative parts, codified as a running commentary in the Midrashim. Interpretation thus gains, maybe for the first time, an independent status, separated from the juristic activity of developing halakha. This in turn enables a development of more explicit and nuanced interpretive tools.

Scholars have reconstructed significant connections between Midrashic techniques and their counterparts in Alexandrian Homeric Scholia, which pre-date them by several hundred years (compare, for example, the rabbinic concept of מקרא עשיר במקומות הרבה “Scripture made rich from [gathering] many places” and Aristarchus’ philological concept of “explaining Homer through Homer”). These techniques are not found, by and large, at Qumran, and so cannot be accounted for as early Jewish traditions. Rather, scholars offered a genetic model, in which Alexandrian methods were adopted by the rabbis, perhaps mediated by Alexandrian Jews.

The rabbis, however, did not develop a systematic theory of interpretation. Their theory must, therefore, be inferred from their practice and terminology. Systematic works on biblical interpretation appear only in the Geonic period, under the clear influence of Muslim scholarship.

In the third century CE the oral Torah was codified in the Mishnah, and the sages in Roman Palestine and Sassanian Babylonia concentrated their effort on explicating this new legal canon. The Palestinian Talmud was redacted much earlier than the Babylonian one (fourth and sixth centuries, respectively) and is much shorter and less developed. The academies in the Galilee, unlike their peers in Babylonia, probably included the whole Mishnah in their curriculum. In their redacted form, however, both exclude those parts that were considered to be irrelevant for their actual lives. Thus the Palestinian Talmud excluded the orders of Kodashim (sacrifices) and Taharot (purities), while the Babylonian Talmud excluded Zeraim (agriculture) and also Taharot (Kodashim was included in the study curriculum in Babylonia, probably because it was considered a substitute to actual Temple ritual).

The two Talmuds use dialectical tools to discuss the Mishnah, and these too are much more developed in the Babylonian Talmud. These dialectical tools are aimed at three goals: 1) to explicate the laconic Mishnah, 2) to harmonize it with other early rabbinic (Tannaitic) traditions, and 3) to align it with the new agendas that developed in the post Mishnaic (Amoraic) study houses, in both halakhic and non-halakhic contexts. Some of the methods developed to explicate the Mishnah are quite similar to those that the early rabbis developed to explicate the Torah, thus manifesting in practice the concept of two Torot, oral and written.

While rabbinic activity largely consisted in the study of texts, this philological activity only came to light through the textualization of these studies in the medieval period. But outside of the rabbinic study house, other sorts of texts were produced. Unfortunately, our knowledge of these texts is limited due to several factors:

(a) Sporadic preservation of texts, due to, among other things, a lack of continuity of local communities (for example, the Jewish community in Alexandria was abolished in the second century CE and restrarted by new immigrants in the fourth or fifth centuries, as is attested in inscriptions)

(b) problems of dating (e.g., the mystical texts named Hekhalot, various parts of which are dated by scholars to diverse periods, from the third to the tenth centuries CE)

(c) the difficulty of distinguishing between Jewish and Christian authorship (e.g., the Latin composition “collatio legum mosaicarum et romanarum,” comparing biblical and Roman laws, written in Rome probably in the fourth century).

In addition to these texts, we also have a significant corpus of inscriptions and mosaics. The large majority of textual activity from this intermediate period is related to magic. Hundreds of incantation bowls written in Aramaic were found in Babylonia. Additionally, amulets written in Hebrew, Aramaic, and Greek have been found in various Jewish communities in and outside Roman Palestine. But only in recent years have scholars begun to see the potential of these amulets and bowls, not only as a window onto Jewish reality and language, but also as an intellectual endeavor in and of itself.


In the Medieval Near East two developments had a major impact on Hebrew philological activities, viz. 1) the textualization in written form of Hebrew oral traditions that had been transmitted for many centuries during the first millennium CE and 2) the adoption of philological and literary models that were current in the contemporary Islamic world. These two developments were intertwined in that the increase in the use of written texts was a phenomenon that embraced the whole of the Islamic Near East, and indeed also the neighboring Byzantine world, in the early Abbasid period (eighth–ninth centuries CE). It is likely to have been brought about, in part at least, by the archival documentary culture of the Abbasid bureaucracy, which developed in the eighth century CE, and the spread in the production of paper in that period.

There were two main clusters of Jewish oral traditions that underwent textualization in the early Islamic period. These were the oral reading traditions of the Hebrew Bible and the oral traditions of Rabbinic literature.

After the destruction of the Second Temple in Jerusalem in 70 CE, the Hebrew Bible was transmitted in written and oral dimensions. Its written form consisted of scrolls written predominantly for liturgical use. Only very few fragments of such scrolls are extant from the period after the destruction of the Second Temple down to the early centuries of Islam. The scrolls that have survived from this period reflect a stabilized consonantal text. The text represented in these written scrolls, however, was read aloud with a pluriformity of reading traditions that did not completely correspond to the written text. The written text acted as the reference point of the reading but was independent of the reading. Some of these oral reading traditions were transmitted in a careful and conservative manner by consecutive generations of teachers and had prestigious authority. Others represented more ad hoc oral performances by individuals. We have only limited access to these oral traditions in the pre-Islamic period. The main sources are the surviving second column of the Hexapla of Origen (c. 185–254 CE), the Latin transcriptions in the Vulgate and writings of Jerome (346-420 CE) and citations of the Hebrew Bible in the text of incantation bowls written in Iraq, many of which appear to be datable to the pre-Islamic centuries of the first millennium CE.

In the early Islamic period this pluriformity of oral traditions and oral culture became textualized by means of the creation of vocalization signs to represent the oral pronunciation of the text and accent signs to represent its musical cantillation. This process of textualization took place in Palestine and Iraq with various degrees of completeness. The most complete textualization was undertaken on a biblical oral reading tradition that had been transmitted carefully for many generations by scholars, known as Masoretes, who were associated with the seat of Jewish authority in Tiberias in Palestine. Known as the Tiberian reading tradition, this was regarded as the most authoritative biblical oral tradition in the Middle Ages. The system of vocalization signs created by the Tiberian Masoretes is known as the Tiberian vocalization. Other systems of vocalization were created for other reading traditions. The next most important was the so-called Babylonian vocalization system. This was used to represent a cluster of traditions of biblical reading that were transmitted in Iraq. There was also another system known as Palestinian vocalization. This was used in biblical manuscripts reflecting for the most part non-authoritative traditions, which can be characterized as ad hoc oral performances by individuals. We also have many biblical manuscripts from the Middle Ages that are vocalized with Tiberian vocalization signs that represent ad hoc oral performances rather than the authoritative Tiberian tradition.

The vocalization signs were not added to biblical texts written in scrolls, the traditional format of manuscript used for Scripture in Judaism, but rather to biblical texts written in codices. One of the Hebrew terms that was used for “Bible codex” in the Middle Ages was מצחף miṣḥaf. This is an Arabic loanword (< Arabic muṣḥaf), reflecting the fact that this form of manuscript was adopted from the surrounding Islamic culture, in which the codex was the normal format for writing the Arabic Qurʾān.

The Tiberian Masoretes were active from the early Islamic period until the end of the tenth century. During this period the Tiberian vocalization signs represented the reading of a living oral tradition. The activities of the Tiberian Masoretes came to an end in the tenth century. Soon after this the Tiberian oral tradition ceased to be transmitted and the primary base of authority began to shift to the written vocalization sign system.

The Tiberian Masoretes undertook other philological activities, in addition to the textualization of the oral reading traditions by vocalization signs. They developed a corpus of Masoretic notes, written in the margins of codices or in independent manuscripts, which were intended to help preserve a stable form of the written text. These notes were in Jewish Palestinian Aramaic, the language of the Jews of Palestine in the Byzantine and early Islamic period. They may have had their origin in oral traditions and so represent another layer of textualization in the early Islamic period.

The Masoretic notes were concerned in principle with the orthography of the written text. The Masoretes also wrote treatises that were intended to systematize rules relating to the vocalization and accent systems. This activity is referred to in some sources as Diqduq, a term that was used in Rabbinic sources to refer to the careful reading or careful study of Scripture. Diqduq in the Masoretic period was concerned in particular with the collation of small differences and with exceptions to general rules in vocalization and accents. The general rules were assumed to be known, or at least would have been assumed to have been taught orally by a teacher and are not always made explicit in the treatises. The earliest treatises of this type, which can be dated to the ninth and early tenth centuries, were written in Hebrew. By the second half of the tenth century Masoretic treatises began to be written in Arabic in Hebrew letters. The field of Diqduq in this period acquired a theoretical apparatus to explain differences in morphological forms of words reflected by their vocalization, which was developed in particular by Karaite scholars in the second half of the tenth century.

A philological activity related to Diqduq was the production of Arabic glossaries and literal Arabic translations of Scripture, which are mainly datable to the tenth century. The Arabic glossaries have come down to us mainly in the form of word lists. The largest surviving corpus of continuous literal Arabic translations was produced by the Karaite Yefet ben ʿEli in the second half of the tenth century.

As remarked, medieval Diqduq was a continuation of Jewish traditions of careful study. In the tenth and early eleventh centuries, however, a new model for the description of the rules of the language of Hebrew Scripture was borrowed by Jewish scholars from the Arabic grammatical tradition. This resulted in the production of systematic grammars of the Hebrew language, written in Arabic, such as the grammar book of Saadya Gaon (first half of the tenth century) and the grammatical works of the Karaite ʾAbū al-Faraj Hārūn (first half of the eleventh century). Both of these scholars, who resided in the Near East, had access to the living Tiberian Masoretic tradition. These grammars differed from the tradition of Diqduq not only in their systematic exposition of grammatical rules, but also in their interest in the phenomenon of human language in general rather than exclusively the elucidation of Hebrew Scripture. In this aspect of their work they were influenced in particular by contemporary traditions of Muslim speculative theology, especially that of the Muʿtazilite movement, which adopted the Aristotelian view of the conventional origin of language. The literary form of these Hebrew grammatical works also followed Arabic models, including, for example, authorial prefaces.

In the tenth and early eleventh centuries a corpus of other types of texts for the study of Hebrew Scripture developed based on intellectual and literary models that were current in the surrounding Arabic Islamic culture. In conformity with their Arabic models, these were written in Arabic. Arabic translations of Scripture were produced that were not literal word-by-word glosses, as was the case with the translations of Yefet ben ʿEli, but rather interpretations of the text expressed in an Arabic literary style that was regarded as felicitous at that period. Such translations were produced by Saadya Gaon in the tenth century and by Karaite scholars, such as Yeshuʿa ben Yehudah, in the eleventh century. In the case of Saadya, his translation was not only free in style but also made adaptations to conform to Rabbanite theological and legal tradition, as did the Aramaic Targums, which Saadya often uses as his source of adaptations.

Closely related to the practice of Arabic Bible translation was the production of complete lexicons of the Hebrew Bible, such as the Egron of Saadya Gaon or Kitāb Jāmiʿ al-ʾAlfāẓ, “The Book of the Collection of Words” of the Karaite David ben Abraham al-Fāsī. These can be regarded as developments of the earlier glossaries, but also were no doubt influenced by the Arabic lexicographical tradition. Some dictionaries of Rabbanite scholars contained words from a corpus of Hebrew wider than that of the Hebrew Bible. This applies in particular to Kitāb al-Ḥāwī, “The Book of Collection” by Hai Gaon (939–1038 CE). The Karaite ʾAbū al-Faraj Hārūn produced a hybrid type of dictionary that offers extensive grammatical analysis of words known as Kitāb Tafsīr al-ʾAlfāẓ, “The Book of the Explanation of Words,” as well as a variety of other related titles.

Another new genre of text that was produced by Jews at this period under the influence of the model of Arabic Qurʾān commentaries was that of the systematic verse-by-verse Arabic commentary of Scripture. These were typically combined with Arabic Bible translations, resulting in a tripartite format of presentation consisting of a verse of the Hebrew Bible, followed by the Arabic translation and then the commentary. Such commentaries typically exhibit awareness not only of the ordered sequentiality of verses but also of the organization of higher units of discourse in biblical books and their literary form. They also adopted various techniques of analysis of textual difficulties from Qurʾān exegesis such as metaphor (majāz) and “virtual underlying form” (taqdīr).

The Jewish scholars who were at the forefront of the adoption of Arabic literary models and intellectual concepts from the Islamic environment were the Rabbanite Saadya Gaon (tenth century) and Karaite scholars in the tenth and eleventh centuries, especially those active in the so-called Karaite school of Jerusalem. The Karaites held that the Bible was the only base of religious authority and rejected the authority of the Rabbinic so-called “oral law” (Mishnah and Talmud). Their philological activities, therefore, focused on the Bible. In the tenth and eleventh centuries these converged particularly closely with the Arabic Islamic environment in form and intellectual framework. They frequently used Arabic script in their Arabic writings, whereas Rabbanite Jews typically wrote Arabic in Hebrew script. The Karaites even wrote some of their Hebrew Bible manuscripts at this period in Arabic transcription, imitating the model of the Arabic Qurʾān.

The other main corpus of Jewish oral traditions that was textualized in the medieval Islamic world was that of Rabbinic “oral law,” at the core of which were the Mishnah and the Talmud. Unlike the biblical oral traditions, Rabbinic traditions did not have the reference point of a written text before they were textualized in written form in the Middle Ages. After their textualization the oral tradition remained as a living reading tradition, as was the case with the biblical oral traditions after their being committed to writing in the form of vocalization systems during the Masoretic period. Analogously to the biblical reading traditions, the Rabbinic reading traditions exhibited a considerable amount of diversity, there being a basic split between the Palestinian tradition and the Babylonian tradition. In addition to stimulating the textualization of the Rabbinic traditions, the Islamic environment provided models of various meta-texts that developed in association with this corpus, such as systematic commentaries and compendia.


When we talk about philology, we focus on textual activity that has two dimensions: the study of texts and the production of texts. With regard to both, we cannot confine ourselves to self-reflective accounts that describe these practices. Instead we extract the practices exemplified in the extant manuscripts and chart the history of the development of philological practices. Our texts do not necessarily reflect on philology, rather they perform it. Indeed, many of these traditions continued down to modern times in both written and oral form. With the rise of modern Hebrew language and literature we see a resurgence of Hebrew in the post medieval environment and the diversification of literary activity.

Geoffrey Khan (PhD, School of Oriental and African Studies, London, 1984) is Regius Professor of Hebrew at the University of Cambridge. His research publications focus on three main fields: Biblical Hebrew language (especially medieval traditions), Neo-Aramaic dialectology, and medieval Arabic documents. He is the general editor of The Encyclopedia of Hebrew Language and Linguistics and is the senior editor of Journal of Semitic Studies. His most recent book is The Tiberian Pronunciation Tradition of Biblical Hebrew, 2 vols, Cambridge Semitic Languages and Cultures 1 (University of Cambridge & Open Book Publishers, 2020).

Hindy Najman (PhD, Harvard University, Near Eastern Languages and Civilizations, 1998) is Oriel and Laing Professor of the Interpretation of Holy Scripture in the University of Oxford and of the Centre for the studies of the Bible at Oriel College. Her research interests encompass Hebrew Bible, Second Temple Judaism, Early Biblical Interpretation, and Scholarly Practices of Reading in Jewish Antiquity. She has written on the Hebrew Bible, the Dead Sea Scrolls, Rabbinic Literature and Pseudepigrapha. Her publications include Losing the Temple and Recovering the Future: An Analysis of 4 Ezra, Past Renewals: Interpretive Authority, Renewed Revelation and the Quest for Perfection, Seconding Sinai: The Development of Mosaic Discourse in Second Temple Judaism, and a recent essay entitled “Ethical Reading: The Transformation of Text and Self.”

Ishay Rosen-Zvi is a full Professor and teaches rabbinic literature in the department of Jewish Philosophy and Talmud at Tel-Aviv University. In 2013 he was elected to the Israel Young Academy of Sciences. He has written on Midrash and Mishnah, as well as on issues of self-formation and collective identity in Second-Temple Judaism and rabbinic literature. His most recent publications are: Goy: Israel’s Others and the Birth of the Gentile (with Adi Ophir) (OUP: Oxford 2018; won the Goshen-Goldstein prize for the best book in Jewish Philosophy for the years 2016-2018); Between Mishnah and Midrash: The Birth of Rabbinic Literature (Open University 2019).