One of the first tasks of the textual critic is to collect the variants among the different witnesses to the text of the Hebrew Bible. This post will introduce some of the Hebrew witnesses to this text.

This is the third in a series of posts on the textual criticism of the Hebrew Bible/Old Testament. Other posts include:

• Textual Criticism of the Hebrew Bible – An Introduction (TCHB 1)
• Textual Criticism of the Hebrew Bible Resources (TCHB 2)

All posts in this series may be viewed here.

Witnesses to the Text of the Hebrew Bible

Extant Hebrew manuscripts and the Hebrew Vorlage that can be retroverted from the extant manuscripts of the ancient versions bear witness to the abstract “text of the Old Testament.” (By “text” I am referring to an abstract concept derived from extant data; by “textual witnesses,” I mean the tangibly different forms of the text; and by “manuscripts, scrolls, and/or codices,” I am referring to the uninterpreted, extant exhibitions of the text.)

The Hebrew Witnesses

It perhaps goes without saying that Hebrew manuscripts are the most important witnesses to the text of the Old Testament/Hebrew Bible for they bear direct witness to it, whereas a retroverted Vorlage, i.e., the text “lying before” a scribe or translator, is always a matter of some conjecture.

There are four Hebrew witnesses to the OT text: the Masoretic text, sometimes called the “received text,” the Samaritan Pentateuch, the scrolls from the Judean desert, and a few, minor additional witnesses.

1. The Masoretic Text

The most important witness to the OT text is the Masoretic text (MT). The Masoretes were a group of Jewish scholars who between 600 and 1000 CE developed a system of notes and signs to preserve the Hebrew text and its reading. The oldest complete manuscript (1008 CE) is the Leningrad Codex B19a (L), which served as the base of BHS and the third edition of BHK (The first two editions of BHK were based on Jacob ben Chayyim’s edition of 1524/25).

The standard critical edition of the Masoretic Text of the Hebrew Bible is Biblia Hebraica Stuttgartensia (BHS). It is available in a variety of formats. I would recommend a hardbound copy if you will be making much use of your Hebrew Bible, though the softbound edition is less expensive and easier to carry around.

• Biblia Hebraica Stuttgartensia – Desktop Version (ed. K. Elliger and W. Rudolph; Deutsche Bibelstiftung, 1967, 1990). This is the large hardbound version. While it has a larger typeface, it is quite bulky. Buy from Amazon.ca | Buy from Amazon.com
• Biblia Hebraica Stuttgartensia – Small Hardcover (ed. K. Elliger and W. Rudolph; Deutsche Bibelstiftung, 1967, 1990). This is a 5″ x 7.5″ hardbound version. It has the benefit of being hardbound, while being a bit smaller than the desk version. Buy from Amazon.ca | Buy from Amazon.com
• Biblia Hebraica Stuttgartensia – Paperback (ed. K. Elliger and W. Rudolph; 5th edition; American Bible Society, 1997). This is a paperback version. It is small and less expensive than the hardbound editions. Buy from Amazon.ca | Buy from Amazon.com
• Biblia Hebraica Leningradensia (ed. Aron Dotan; Hendrickson Publishers, 2001). This is a thoroughly revised, reset, and redesigned — and the most accurate — edition of the Leningrad Codex in print. Buy from Amazon.ca | Buy from Amazon.com
• Biblia Hebraica Stuttgartensia with Greek New Testament (American Bible Society, 1996). Ideal for those who want a complete Christian Bible with both the Hebrew Old Testament and the Greek New Testament. Buy from Amazon.ca | Buy from Amazon.com
• Hebrew-English Tanakh (Student edition; Jewish Publication Society of America, 2000). While this is not an edition of BHS (and therefore has no critical apparatus), I note it because it may be useful for beginning students. It has the Hebrew text on one side of the page and the Jewish Publication Society’s English translation on the other. Buy from Amazon.ca | Buy from Amazon.com
• David Noel Freedman, Astrid B. Beck, James A. Sanders (Eds.), Leningrad Codex: A Facsimile Edition (Eerdmans , 1998). If you want to impress your professor (or students), then this is the Hebrew Bible for you! This is a facsimile version of the Leningrad Codex, the oldest complete copy of the Hebrew Bible (the name is due to the fact that it was in a museum in Leningrad, when it was Leningrad). This photo-plate edition of the entire text (black-and-white high resolution plates, with additional full-colour plates of carpet pages and sample text pages) is beautifully produced and bound. Buy from Amazon.ca | Buy from Amazon.com

Another significant witness to the Masoretic text is the Aleppo Codex, which is thought to date earlier than the Leningrad Codex. The codex was written by Shelomo ben Baya’a, but according to its colophon it was pointed by none other than Moses ben Asher (930 CE). It was reported to be destroyed in a fire in 1948; but as it turned out, only the Torah portion was lost while the other books were saved (thus while it is the earliest extant copy of the Masoretic text, it is incomplete). The codex has now been photographed and is the basis for the Hebrew University Bible Project (HUBP), of which the books of Isaiah, Ezekiel, and Jeremiah have been published:

• H. Goshen-Gottstein, ed., The Book of Isaiah: The Hebrew University Bible [Hebrew University Bible Project; Brill, 1997] Buy from Amazon.ca | Buy from Amazon.com).
• C. Rabin, S. Talmon, and E. Tov eds., The Book of Jeremiah: The Hebrew University Bible [Hebrew University Bible Project; Magnes Press: The Hebrew University, Jerusalem, 1997] Buy from Amazon.ca | Buy from Amazon.com).
• Moshe H. Goshen-Gottstein and Shemaryahu Talmon, eds., The Book of Ezekiel: The Hebrew University Bible [Hebrew University Bible Project; Magnes Press: The Hebrew University, Jerusalem, 2004 ] Buy from Amazon.ca | Buy from Amazon.com).

The term “Masoretic text” is an abstract term for the distinctive kind of text these scholars produced. As a rule the term is restricted to the final form of that text, a manuscript produced in the tenth century by Aaron ben Asher, the primary base for the hundreds of medieval manuscripts. It is now known that some of the oldest DSS reflect the essentially same text inherited by the Masoretes, and so called the proto-Masoretic text (proto-MT).

The sequence of books in the MT differs from that of the Septuagint codices on which the order of our English Bibles is based. The former’s tripartite division is: The Torah, the Prophets, and the Writings; while the fourfold division of the latter is: the Law, historical books, poetic and wisdom books, and prophetic books.

The activity of the Masoretes was fourfold. First, the Masoretes “hedged in” the consonantal text they inherited with a Masorah, their scribal notes in the margins around it. Earlier scribes had settled upon that text type by the end of the first century CE. Scribal precision in transmitting the consonants before the activities of the Masoretes is reflected in the Talmud. R. Ishmael said: “My son, be careful, because your work is the work of heaven; should you omit (even) one letter or add (even) one letter, the whole world would be destroyed” (b. Sota 20a). Generations of Masoretes contributed an apparatus of instructions written in the margins around the text. By so hedging the text the Masoretes hoped to assure its precise transmission even to its smallest details.

Second, above and below the inherited consonants the Masoretes added vocalization — vowel points to preserve its accompanying oral tradition. Prior to the Masoretes, scribes more or less sparingly represented important vowels by four Hebrew consonants: ×™ (y), ו (v), ×” (h), and ×? (’) called matres lectiones (“mothers of reading”). Vowels, of course, can be decisive in meaning. Contrast the difference vowels make with the consonants, “fr”: “far,” “fir,” “fire,” “for,” “fore,” and “fur.” A story told in the Talmud illustrates that the scribes recognized the importance of an accurate oral tradition. In the story we are told David reprimanded Joab when he killed only the men of Amalek and not the “remembrance” (zeker) of Amalek. Joab, however, defended himself, noting his teacher taught him to read “to kill all their ‘males'” (zakar). Later, however, Joab drew his sword against his poor teacher who taught him incorrectly (b. B. Bathra 21a-b).

Third, the Masoretes added a system of accentuation to the text. These diacritical accents that signify the melodious chant serve to beautify and to add dignity to the reading of the text, to denote the stress of the word, which can be as meaningful as the difference between the English “pres´-ent” and “pre-sent´”, and to denote the syntactical relation between words as either conjunctive or disjunctive. For instance, it makes some difference where one places the accents in Isaiah 40:3:

The voice of him that crieth in the wilderness, Prepare… (KJV).
A voice of one calling: “In the desert prepare…” (NIV, cf. the footnote).

Fourth, the Masoretes also preserved various para-textual elements — the verse and paragraph divisions of the text found in the oldest manuscripts and ancient textual corrections. The numbering of the verses and division of the books into chapters, however, was done in the Latin Vulgate, not in Jewish sources. In addition, the Masoretes either preserved and/or added corrections to the received text by marks within the text and within the margin. In the MT one finds inverted nuns (see before and after Num 10:35-36 and Ps 107:23-18), looking something like half-brackets, and extraordinary points, among others signals, to call attention to the received consonants in need of correction. For instance, the Sebirin (סביר) which in about 350 instances introduces a marginal note to an unusual word and proceeds to give the usual form of the expected expression.

The most important corrections are the Ketiv-Qere (K-Q) variants. Ketiv (= “written”) refers to the consonants in the text, for which the reader must guess the vowels, and the Qere (= “read”) to consonants in the margin to which the reader must add the vowels found in the text. In more than 1300 instances there are two readings (one written, the Ketiv; and one read via the vowel points, the Qere) given as at times the text was felt to be unsatisfactory on grammatical, esthetic, doctrinal, or textual grounds. At first the Qere readings were optional corrections of the consonantal text, but by the time of the Masoretes they had become obligatory. Some other interesting changes include the tiqqune sopherim (scribal corrections) and the itture sopherim, scribal omissions. The former include unseemly references to God; the latter include various grammatical omissions (see Gen 18:22 or 1Sam 3:13).

For those interested in learning the basics of the Masorah of BHS (i.e., the Masoretic scribal notes in the margins of the Hebrew Bible), the following works are quite helpful:

• Page H. Kelley, Daniel S. Mynatt, and Timothy G. Crawford, The Masorah of Biblia Hebraica Stuttgartensia: Introduction and Annotated Glossary (Grand Rapids: Eerdmans, 1998). Buy from Amazon.ca | Buy from Amazon.com

• Israel Yeivin, Introduction to the Tiberian Masorah (Trans. and ed. by E.J. Revell; SBLMS 5; Scholars Press, 1985). Buy from Amazon.ca | Buy from Amazon.com

2. The Samaritan Pentateuch

The second major witness to the Hebrew text of the OT is the Samaritan Pentateuch (SP). The Samaritans, once a very large sect, are a now small group still centered at modern Nablus, biblical Shechem/Sychar. Most Christians know this sect from Jesus’ famous conversation with the Samaritan woman (John 4). As that story shows, the Samaritans distinguished themselves from Judaism by their worship on Mount Gerizim, not Jerusalem. They restricted their Bible to the Torah because Moses, its traditional author, only called for a central sanctuary without designating a specific location. In the Prophets, however, David selected Jerusalem as the central sanctuary, and the Hagiographa celebrates that city.

The SP began its own history in the last quarter of the second century BCE, though the sect itself may be centuries older. The SP differs from the MT in some 6000 instances. While it is true that a great number of these variants are merely orthographic and trivial, it is significant that in about 1600 instances the SP agrees with the LXX against the MT. To be sure, some of these variants are tendentious, such as modifications to the ten commandments noted below. Others, however, demonstrate a tendency toward expansion. Basing himself on Gesenius, the first to classify the variants between the SP and the MT in a thorough and convincing way, Waltke demonstrated from recent philological and textual research that the SP, which is written in a special version of the “early” Hebrew script, presents a secondarily modernized, smoothed over, and expanded text (B. K. Waltke, “The Samaritan Pentateuch and the Text of the Old Testament,” in New Perspectives on the Old Testament [ed. J. B. Payne; Waco, TX: Word, 1970] 212-239; Buy from Amazon.ca | Buy from Amazon.com).

It is now known from the DSS that the Samaritans adapted a pre-Samaritan, Jewish text to their idiosyncratic theology. For example, they were able to make the worship on Mount Gerizim the tenth commandment by combining the first two commandments into one and by adding texts from Deut 11:29a, 27:2b-3a, 28:4-7, and 11:30 after Exod 20:17, numbering the material from Deut 28:4-7; 11:30 as the tenth commandment. (In these roughly added interpolations from Deuteronomy into Exodus, it is instructive to note, the change in divine names, doublets, change of style, and change of vocabulary, are akin to the criteria by which literary critics historically identified sources in the Pentateuch.)

In the light of Qumran, the SP has become a very important witness to a form of the Hebrew text that once enjoyed use, as shown by its agreements with the Qumran texts, the LXX, the NT, and other Hebrew texts. Indeed, because of the sectarian character of the Samaritans, the SP gives us a Hebrew witness independent of the changes that developed in mainline Jewish transmission, at least after about 100 BCE.

3. The Dead Sea Scrolls

The DSS are copied in Hebrew, Aramaic, and Greek. By the techniques of palaeography, numismatics and archaeology, they are dated from mid-third century BCE to 135 CE. Most manuscripts were found in the eleven caves in the mountains just west of Khirbet Qumran (15 km south of Jericho near the Dead Sea), which ceased to exist after 68 CE. These caves yielded some 800 scrolls of all the books of the Bible, except Esther. The other principal sites, Nahal Hever and Wadi Muraba’at, yielded texts that are somewhat later, all of which belong to the proto-MT. The scrolls found at Masada, which fell to the Romans in 70 CE, are also proto-MT. While there are dangers of reading post-70 CE realities back into these biblical texts, the best classification of these scrolls is that offered by Tov, who divides them into four different text types:

1. First, there are what is called the Proto-Masoretic texts. As noted, the Masoretes finalized the proto-MT. The great number of Qumran scrolls belonging to this text type, about 47% of them, may reflect their authoritative status.
2. Second, there are Pre-Samaritan texts (ca. 2.5%). As mentioned above, the Samaritans adopted and adapted an earlier Jewish text type attested at Qumran. These scrolls have the characteristic features of the SP, aside from the thin layer of ideological and phonological changes the Samaritans added. Because of this difference, however, Tov is right in calling this text pre-Samaritan, rather than proto-Samaritan as has been the custom. This text type is at least as old as Chronicles, for where Chronicles (ca. 400 BCE?) is synoptic with Genesis, it displays a text type like these texts, not like the MT. Since this text type was modernized by at least 400 BCE, the archaic proto-MT of the Pentateuch, and so the Pentateuch itself, must be much older.
3. Third, there are Septuagintal texts among the DSS. The original Greek translations of certain books of the OT were based on a distinctive text type. Some Qumran Hebrew scrolls, most notably 4QJer^{b, d}, bear a strong resemblance to the Septuagint’s Vorlage. The Septuagintal text type comprises about 3.5% of the Qumran biblical texts.
4. Fourth, a large number of Qumran scrolls (ca. 47%) are not exclusively close to any one of the types mentioned above and therefore classified as non-aligned. Tov explains: “they agree, sometimes significantly, with MT against the other texts, or with SP and/or LXX against the other texts, but the non-aligned texts also disagree with the other texts to the same extent. They furthermore contain readings not known from one of the other texts” (Tov, Textual Criticism, 116).

In addition to these text types, Tov also identifies a group of texts that reflect a distinctive orthography (i.e. spelling, similar to English “favor” versus “favour”), morphology, and free approach to the biblical text visible in content adaptations, in frequent errors, in numerous corrections, and sometimes, also, in negligent script. Tov thinks that only these scrolls were produced in or around Qumran, and therefore describes them as written in the “Qumran Practice.”

For more information on the Dead Sea Scrolls, including a full listing of the Critical Editions of the Dead Sea Scrolls, see my Dead Sea Scrolls Resources page.

4. Additional Hebrew Witnesses

The oldest evidence to the Hebrew Bible are two minute silver rolls about the size of cigarette butts that could be worn around the neck found at Ketef Hinnom in Jerusalem. They contain the priestly blessing (Num 6:24-26) in a slightly different formulation than MT and are dated to the seventh or sixth century BCE. Some other witnesses include the so-called Nash Papyrus (second century BCE), containing a liturgical text of the Decalogue (Exodus 20 and Deuteronomy 5) and the Shema (Deut 6:4-5). There are also many fragments of biblical texts contained in mezuzot, head-tefillin, and arm-tefillin from the second and fist centuries BCE until the first and second centuries CE. These often differ from the MT, possibly because they were written from memory as implied in the Talmud.