A court of law relies on witnesses to establish the facts of a
case. But for those who seek the “facts” of the original biblical texts, no
firsthand witnesses exist. We have only the testimony of various manuscripts,
produced hundreds of years after the Bible's books were completed. And even if
we had an autograph copy of, say, the Book of Ezra, it would not answer all our
questions, for it was created at a time (2400 years ago) when writing was
imprecise--even before the invention of punctuation.
Preface to the 1999 Hebrew-English Edition"
Through the intervening centuries, scribes have figured out how
to record the oral traditions more precisely. At the same time, during each
transmission of the books from person to person, uncertainty has grown. For
schools have sometimes disagreed on pronunciation. Handwriting has not always
been legible. And every scribe has occasionally made mistakes in copying.
Witnesses testifying in court often disagree. Little surprise,
then, that the Bible's textual “witnesses”--farther removed from the original
“event”—differ from each other in a wide range of small ways: spelling,
punctuation, layout of poetry, and so on. Sometimes entire verses appear in only
a few manuscripts.
So which version is true? This was the first question faced
by the [editors] in preparing [the] Hebrew text.
The Unbroken Chain of Uncertainty
Accuracy has been ensured via side documentation—part of what is
called masorah. This gives rise to a “masoretic text”—a Bible that
accords with the masorah. Yet the number of details has been too vast for
masorah. to address all room for disagreement. And the masoretic notes
have been neglected through the ages; written in shorthand, they are often
vague, and their own textual witnesses sometimes disagree.
Much masorah seems to have been created only after
problems arose; in such cases, it could only reinforce the torn textual fabric,
not mend the hole. Unable or unwilling to choose between variants, scribes
sometimes preserved two versions of a word side by side—transmitting both.
Furthermore, by nature the Bible is not predictable. Because of
its spiritual subject matter, its choice of words must be improbable at times.
So is a given puzzling phrase due to scribal error—or religious mystery? When to
expect the text to follow rules of grammar—and when to allow for artistic
Despite these pitfalls, Bible scholars have always refined the
text as they found it. Each expert begins with a different set of available
manuscripts, from scribes of varying (and uncertain) reliability. They each use
different methods for resolving textual problems. So the experts come to
different conclusions as to what is the “best” Bible text.
Ironically, the result is a Bible whose text continually
evolves—the changes being justified to preserve the accuracy of tradition.
Thus, an early printed Bible edited in Italy by R. Jacob ben Hayyim Ibn-Adoniyah
(1525 C.E.) reconstructed the work of the Tiberian textual tradition from six
hundred years earlier. His effort was impressive enough that soon afterward,
owners of old manuscripts all across Europe altered their parchments to match
his newly authoritative book.
As mistakes were corrected, new ones appeared. R. Meir Letteris
of Austria edited a Hebrew Bible first published in 1852 C.E., based on lists of
“corrections” by experts who perceived mistakes in earlier editions. It became
the standard Hebrew text among many Jews to this day. Yet like all prior printed
Bibles, it contained hundreds of its own typographical errors.
Meanwhile, in the modern era, certain early medieval
manuscripts—safeguarded in isolated Middle Eastern communities—were brought to
the attention of Bible scholars. These have proven to be the oldest known
witnesses of the now-standard Tiberian tradition. Only recently has the wider,
evolutionary, corrective process taken these unusually reliable texts into
account. These manuscripts still contain inconsistencies and differ slightly
from each other. But on the whole, they confirm the “received” (evolutionary
European) traditions of the Bible text, especially for the Torah.
Amazingly, manuscript differences are truly minor. More than
99.9 percent of the time, the masoretic Bible's witnesses give identical
accounts. Rarely does the variation impact the meaning of a given verse. Yet
even this small uncertainty can vex a publisher who—for each detail of the
text—must pick one possibility over another.
If a definitive text of the Hebrew Bible does not exist, the
best a publisher can do is produce a defensible text that is sufficiently
accurate for the edition's intended purpose. Therefore [we] explain our
text's history and [the editing] approach [used]—and let the reader judge its
The History of Our Hebrew Text
Since ancient times, Jews have traced the chain of transmission
of Scripture: “Moses received Torah at Sinai and handed it on to Joshua, Joshua
to the elders, and the elders to the prophets. And the prophets handed it on to
the Great Assembly. . .” (Mishnah Avot 1:1). For the present volume, the
textual transmission history is as follows.
Aaron ben Moses Ben-Asher (Tiberias, c. 930 C.E.)
An industrious family of masoretes once lived in the Galilean
town of Tiberias (an ancient center of Jewish scholarship). The last in their
line of scholars was Aaron ben Moses Ben-Asher, who flourished circa 930 C.E. He
authored a classic masoretic treatise. He is the first known scribe to complete
a manuscript of the entire Bible (whose books had been preserved somewhat
independently). An important part of his work included the proofreading of
others' manuscripts, which is how he enters into our picture.
Samuel ben Jacob (Egypt, 1010 C.E.)
Two generations later, a scribe in Fostat (Egypt's thriving
center of trade and learning) spent years composing a Bible codex. Noting
its completion in 4770 A.M. (1010 C.E.), he recorded that he copied from several
manuscripts into this one volume: “Samuel ben Jacob wrote out the consonants,
vowels, punctuation, accents, and annotations of this codex of Scripture from
the texts checked and corrected by the late master Aaron ben Moses Ben-Asher; it
has been checked and corrected per tradition.”
Samuel's Bible contains sixty thousand marginal notes on the
text, including more than a thousand divergences between consonantal text (kethib)
and reading tradition (qere).
In proofreading and correcting his work, Samuel ben Jacob missed
(or let stand) hundreds of errors—which is actually an impressive result, given
the millions of characters in a Hebrew Bible. As the contemporary scholar E.J.
Revell comments, “This is a long way from perfection, but it is close to ideal
when compared to the situation in most [later] medieval manuscripts.”
Recent Editions of the Leningrad Codex
Today, Samuel ben Jacob's work is the oldest known complete
Hebrew Bible, and the oldest complete representative of the Ben-Asher tradition.
For centuries, however, it was kept out of circulation, unknown to historians or
Bible editors. Then in 1840 C.E., a manuscript collector announced possession of
this Bible—which has since become known as the Leningrad Codex.
Repeatedly since then, international teams of Christian and
Jewish scholars (both religious and academic) have edited this codex for modern
use. The first group, led by Rudolph Kittel and Paul Kahle, made it the base
text for a critical edition, Biblia Hebraica Kittel (BHK, 1937). After
World War II, another team revised BHK, producing Biblia Hebraica
Stuttgartensia (BHS, 1967–1977), for which Karl Elliger, Wilhelm Rudolph,
and Gerard Weil served as lead editors. Then, at the University of Michigan, H.
Van Dyke Parunak and Robert Eckert devised computer-readable codes for the
biblical text's characters and main features; Parunak oversaw the transcription
of BHS into three megabytes of data (1982). Soon thereafter, Richard E. Whitaker
of the Claremont Graduate Schools coordinated revisions. Finally, J. Alan Groves
of Westminster Theological Seminary (Philadelphia) with Emanuel Tov of The
Hebrew University (Jerusalem) directed a proofreading team (1987), a project
that JPS helped to fund.
The result is called the Michigan-Claremont-Westminster (MCW)
electronic BHS. It has provided JPS with a text nearly identical to the
Leningrad Codex manuscript. Each round of revision has corrected previous
typographic errors and misreadings while introducing a smaller number of other
typos and mistakes. Its machine-readable text format has nearly precluded new
typos in our own production process. Meanwhile, BHS notes have provided vital
supporting documentation. . .
-- Rabbi David E. Sulomm Stein, Managing Editor
The JPS Translation of the Holy Scriptures
The translation of Tanakh, the Holy Scriptures, produced
by the Jewish Publication Society, was made directly from the traditional Hebrew
text into the idiom of modern English. It represents the collaboration of
academic scholars with rabbis from the three largest branches of organized
Jewish religious life in America: Reform, Conservative, and Orthodox. Begun in
1955, the ongoing translation was published in three main stages: The Torah
in 1962, The Prophets (Nevi'im) in 1978, and The Writings (Kethuvim)
in 1982. These three volumes, with revisions, are now brought together in
a complete English Tanakh (Torah-Nevi'im-Kethuvim), the latest link in
the chain of Jewish Bible translations.
-- The Jewish Publication Society
To aid modern readers, its various editors have brought the
codex's outward features in line with the more familiar evolution of Bibles, by:
Adding chapter and verse numbers (invented after the codex was
Changing the order of books (putting Chronicles at the end,
rather than before Psalms);
Redividing Psalms to show 150 chapters rather than the codex's
Adding end-of-verse punctuation where Samuel ben Jacob had
Inserting typographic markers to show the codex's
Placing each qere entry in the text (rather than in the
margin), following its kethib, and transferring pointing to the qere
consonants—which occasionally meant inferring pointing (a dagesh or
maqqef ) not in the codex;
Omitting the diacritical sign rafé (ubiquitous in the
codex) in all but six places, which most modern Bibles have dropped as
superfluous and hard to print;
Turning many marginal notes on superfluous letters (yatér)
into qere entries;
Tripling the number of masoretic notes by filling in
Flagging the apparent scribal errors.