When the Dead Sea Scrolls were written, no canonical Bible existed. That is, in the two or three centuries before the Roman destruction of the Temple in 70 A.D, there was no one list of sacred books that was considered authoritative. At the same time, there was no clear border between biblical books and nonbiblical books. Rather, different groups of Jews considered different books authoritative, even though all Jews accepted the Torah, or Pentateuch—that is, Genesis, Exodus, Leviticus, Numbers and Deuteronomy. The Torah was, after all, the source of the Law, which provided the underpinning of Jewish ritual and daily life.
But the Dead Sea Scrolls reveal a surprising fact: Even in the case of the Torah, there was no fixed text either of the Torah as a whole, or of any of the individual books. Among the scrolls is a whole group of texts that are related to, but differ from, the present-day books of the canonical Torah. Some of the texts are simply copies of biblical books with variants, the result of centuries of hand copying (scribal error or manipulation) and textual growth. These documents provide critical new material to the text critic who attempts to recover the best text of a biblical book, using all copies available.
Some of these texts, however, differ markedly—at times startlingly—from the standard authoritative Jewish version of the Bible, known as the Masoretic text, or MT for short. Nor do they resemble the two other major biblical textual traditions, the Septuagint (or LXX for short) and the Samaritan Pentateuch. The Septuagint is a Greek translation made for the Jews of Alexandria, Egypt, the first five books of which were translated in the third century B.C. from a Hebrew text that differs somewhat from MT. (According to legend, the name Septuagint, which comes from the Latin term for “seventy,” refers to the 72 Jewish translators brought to Egypt by Ptolemy Philadelphus [285–246 B.C.] to translate the Torah.) More about the Samaritan Pentateuch 036later. Suffice it to say that MT is the authoritative text for Jews and Protestants; LXX, for the Orthodox churches; and the Samaritan Pentateuch, for the small group of Samaritans who still live in Nablus and a few places in Israel. Each of these traditions is represented in various fragmentary manuscripts of the Pentateuch found among the Dead Sea Scrolls.
But some of the seemingly biblical manuscripts from Qumran differ considerably from all of these traditions. The question I would raise is, In ancient times, how far could these texts deviate and still be considered biblical? Or authoritative? Scholars themselves are somewhat unsure, calling them “parabiblical” or “quasibiblical.” Those terms, however, describe the texts only from our viewpoint. To us, they are not canonical and therefore cannot be biblical. But to the people who copied and read them two thousand years ago, they may have been just as authoritative as the texts we consider biblical today.
Let’s look at a few parabiblical texts:
Our first example is a Dead Sea Scroll called 4QDeuteronomyn, which was copied in the late first century B.C. and which contains a text of the Ten Commandments.1 (The “4Q” that appears so often in Dead Sea Scroll designations stands for Qumran Cave 4, where more than five hundred different manuscripts, all fragmentary, were found.) The Ten Commandments appear in two places in our canonical Torah—in Exodus 20 and in Deuteronomy 5, but the two versions are not exactly the same (see sidebar to this article). The fourth commandment in Exodus (Exodus 20:8–11) bids the Israelites to “remember” the Sabbath day. In Deuteronomy (Deuteronomy 5:12–15), however, the Israelites are commanded to “observe” the Sabbath 037day. And that’s not all: The rationales differ in Exodus and Deuteronomy. In Exodus, the Israelites must remember the Sabbath because the Lord rested on the seventh day after creating the universe in six days. In Deuteronomy, the reason given is that they were slaves in Egypt.
The two different versions were already well established by the time the Dead Sea Scrolls were copied. How do we know this? Because both are referred to in 4QDeuteronomyn. As shown in the sidebar to this article, 4QDeuteronomyn presents yet another version of this commandment. Here the scribe begins with the Deuteronomy version (“Observe the Sabbath day”) and gives Deuteronomy’s reasoning: “And remember that you were a servant in the Land of Egypt…therefore the Lord your God has commanded you to observe the Sabbath day to sanctify it.” But the text doesn’t stop there. Rather, it picks up with the justification given in Exodus: “For six days the Lord made the heavens and the earth, the sea…and he rested on the seventh day; therefore the Lord blessed the Sabbath day to sanctify it.” The scribe has smoothed out, or harmonized, the two texts by combining both justifications into one (very long!) Sabbath commandment.2
What did first-century readers think when encountering this text? We can only speculate, but probably they would have recognized it as a harmonization of the other two existing versions. After all, at the time there probably were other manuscripts of Exodus and Deuteronomy that contained the Ten Commandments in the versions with which we are familiar today.a So a careful reader would have recognized that a change had been made. But would that have made any difference to the authority of the text? Probably not. This seems to be a major difference between Second Temple period Jews (living before the Roman destruction of 70 A.D.) and the modern Jewish or Christian reader.
For Second Temple period Jews, the authority of these books lay in each book’s general message rather than in its precise words or their order.3 The words of the biblical text could be manipulated—moved around, updated, added to—without detracting from the authoritative status of the book. This may not have been true for all Jews in the Second Temple period, but it certainly seems to have been true for the Jews of Palestine.
Thus, in 4QDeuteronomyn the important point is the command to observe the Sabbath, which is unchanging; importing text from Exodus into the passage in Deuteronomy simply adds weight to the commandment.
How far could this process of manipulation go before a biblical book was so modified that it became another edition of the same book, or an entirely different book? Did books lose authority in the process?
Consider another example. I have already mentioned three different versions of the Pentateuch: the Masoretic text, the Septuagint and the Samaritan Pentateuch. The Masoretic text is actually a medieval text, but it is based on manuscripts at least as old as the Dead Sea Scrolls; prototypes of the Masoretic text (so-called proto-Masoretic or proto-Rabbinic texts) have been found at Qumran. So have parts of various Hebrew base texts from which the Septuagint was translated. These are sometimes called “proto-Septuagintal.” The Samaritan Pentateuch is a harmonized text like 4QDeuteronomyn, as illustrated by the following example:
In the standard biblical text of Genesis 31:4–13, Jacob, who is still living with his father-in-law, Laban, tells his wives, Rachel and Leah, about a dream he had long before, in which God commanded him to leave Aram and return to Canaan. “During the mating of the flocks,” Jacob rather abruptly recalls, “I once had a dream,” in which a messenger of God told him “to arise, go forth from this land and return to the land of your fathers.” This is the first we’ve heard of this dream. Did Jacob simply make it up to justify the pending departure to his wives? The Samaritan Pentateuch provides the answer. Jacob’s dream is first described in detail when he dreams it (after Genesis 30:36) and then later is repeated to his wives. The insertion reads as follows:
And the messenger of God spoke to Jacob in a dream, and he said “Jacob!” And he said “Here I am.” And he said, “Raise your eyes and see all the he-goats climbing upon the flock, striped, speckled, tan and spotted. For I have seen everything which Laban has done to you. I am the God of Bethel, where you anointed a pillar and vowed a vow to me. And now, arise, go forth from this land and return to the land of your fathers and I will deal well with you.”
The language of the dream is exactly the same as Jacob’s later report to his wives. Any doubts about Jacob’s veracity in recounting the dream are laid to rest by including it in the biblical text.
This type of harmonization was used not just once but systematically throughout the entire Samaritan Pentateuch. When the Samaritan community adopted this edition as their canonical Torah, they also made certain sectarian changes. Instead of including the veiled references to Jerusalem and Mt. Zion that appear in MT, the Samaritan Pentateuch refers to Shechem and Mt. Gerizim, the Samaritan holy mountain. These are the places that God “has chosen,” the Samaritan Pentateuch says. In MT, the parallel passages refer (obliquely) to Jerusalem and Mt. Zion as the places that God “will choose.” But before these sectarian changes were made and this edition was adopted by the Samaritan community, the harmonized Samaritan Pentateuch, was in general circulation in Palestine. This is obvious from the fact that several copies of “proto-Samaritan” texts have been found at Qumran (for example, 4QNumb and 4QpaleoExodm). At this time, and at least for some groups of Jews, there was simply no distinction between proto-Samaritan texts and proto-Masoretic texts; they were just different copies of the same authoritative book.
For the Sabbath commandment and the story of Jacob’s dream, the scribal manipulations were extensive, but the passages are still recognizable as constituting the same biblical text. These are true harmonizations, which smooth over bumps in the text but do not add anything new. This kind of change does not seem to have altered the book’s authority in any way. What would happen, though, if something brand new was added to the text?
This is just what occurs in several other Dead Sea Scroll manuscripts that are referred to as parabiblical 050or quasibiblical. Let us look at an example from one of these parabiblical texts, called 4QReworked Pentateuch. The complete document probably included the entire Pentateuch on one large scroll. Five manuscripts of the Reworked Pentateuch have been preserved, only one of which—designated 4Q365—we will look at here.4 From the shape and form of the letters, 4Q365 can be dated to about 75 B.C.
As the title suggests, the scribe has reworked or changed the biblical text to a greater extent than we have seen with the Sabbath commandment or Jacob’s dream. One type of change is the addition of brand new material. In 4Q365, a substantial addition was made to the Song of Miriam. In the standard biblical text, the Song of Miriam appears immediately following Moses’ victory song at the Reed Sea (Exodus 15:1–18) and consists of just one sentence. Exodus 15:20–21 states: “Then the prophet Miriam, Aaron’s sister, took a tambourine in her hand; and all the women went out after her with tambourines and with dancing. And Miriam sang to them: ‘Sing to the Lord, for he has triumphed gloriously; horse and rider he has thrown into the sea.’” Not only is it short, it is simply a repetition of the first line of Moses’ song in Exodus 15:1.b
But is that all that Miriam sang?
The answer in the Reworked Pentateuch is a resounding “No!” The additional text is fragmentary, but there is no doubt that it belongs here. After what we know as Exodus 15:21 and before Exodus 15:22, 4Q365 inserts seven additional lines to Miriam’s song. The fragmentary addition goes like this:
1. You despised[c 2. for the majesty of[ 3. You are great, a deliverer[ 4. The hope of the enemy has perished, and he is for[gotten (or: has cea[sed) 5. they perished in the mighty waters, the enemy[ 6. Extol the one who raises up, [a r]ansom…you gave[ 7. [the one who do]es gloriously[
Miriam’s song is addressed to God and celebrates his victory over his enemies at the sea.d The additional text closes a perceived gap in the text and adds to the drama of the narrative, extending the theological message: God is the victor at the Reed Sea—not Moses or the Israelites. It interprets the text from within the text, making sure the reader understands the meaning of the story.5 Thus the scribe was doing his job as the keeper of the tradition, making sure that the message was heard and properly understood by each new generation. This was not meant to be blasphemous or false. In fact, it was in the very act of reworking the text that the scribe indicated just how sacred or important the text actually was. An unimportant text would be discarded or forgotten; a sacred text like the Pentateuch, however, was constantly shaped and reshaped by generations of scribes and interpreters.
Would the first-century reader have recognized that this was an altered text of Exodus? Probably. Would that have affected the authority of this Exodus text? We 051simply don’t know. If it was considered acceptable to manipulate words in a text, as in the treatment of the Sabbath commandment in 4QDeuteronomyn or the addition to Jacob’s dream in the Samaritan Pentateuch, why wouldn’t the type of change in 4Q365—the addition to Miriam’s song—be equally acceptable? If it were, 4Q365 would be just another manuscript of the Torah, equal in authority to any other manuscript.6
But I am not so sure. Two pieces of circumstantial evidence give me pause. First, this addition to Miriam’s song did not continue to be copied in the late Second Temple period. Eventually it fell out of general circulation (at least we have no evidence of its continued use). Second, the unique passages in 4Q365 are not quoted elsewhere in Second Temple literature, with one exception.7 Further, the vast body of rabbinic literature, for example, knows nothing of it. Clearly, 4Q365 was not widely known, certainly not beyond its own life as a manuscript. For that reason, I am inclined to think that while 4Q365 may have had authority for a limited audience around the time of its production, it was never generally accepted as authoritative.
By the end of the first century A.D., we begin to see some changes in the notion of both a canon, or list of sacred books, and an authoritative, unchangeable text. Josephus, the first-century Jewish historian, mentions a list of 22 sacred books (Contra Apion 1.37–43). His list implies that whatever is not included is not sacred. Similarly, 2 Esdras 14:45 talks about 24 “public” books that were written by divine inspiration (in addition to 70 “hidden” books known only to the wise).
With regard to fixation of the text: A number of fragmentary biblical scrolls dating to the second century A.D., discovered in caves south of the Wadi Qumran, suggest that at about the time a canon was developing, so too was the notion of a fixed authoritative text. All of the second-century A.D. biblical manuscripts from the caves south of Qumran are proto-Masoretic texts; by this period, other text types seem to have fallen out of circulation. Thus, after the fall of the Temple in 70 A.D., the canonical list becomes fixed in Palestinian Judaism, as does the text of those canonical books.8 No deliberate changes would henceforth be made. A great tradition of innerbiblical exegesis—so clearly reflected in the Dead Sea Scrolls—had come to an end.
We like to think of Holy Writ as unchanging, but the ancients didn’t. A study of the Dead Sea Scrolls reveals that texts could exist in different forms—even be consciously modified—without losing their sanctity.
You have already read your free article for this month. Please join the BAS Library or become an All Access member of BAS to gain full access to this article and so much more.
I say “probably” because, unfortunately, the Sabbath commandment has not been preserved in any other Qumran manuscript of Exodus or Deuteronomy, although it does appear in the phylactery (tephillin) texts.
Sidnie White Crawford, “4QDeutn,” in Discoveries in the Judaean Desert XIV, ed. Eugene Ulrich et al. (Oxford: Clarendon Press, 1995), pp. 117–128, pls. 28–29.
2.
Emanuel Tov, “The Nature and Background of Harmonizations in Biblical Manuscripts,” Journal for the Study of the Old Testament 31 (1985), pp. 3–29.
3.
Ulrich, “The Bible in the Making: The Scriptures at Qumran,” in The Community of the Renewed Covenant, ed. Ulrich and James VanderKam (Notre Dame, IN: Univ. of Notre Dame, 1996), p. 84.
4.
Tov and Sidnie White (Crawford), “Reworked Pentaentateuch,” in Discoveries in the Judaean Desert XIII, ed. VanderKam et al. (Oxford: Clarendon Press, 1994), pp. 187–352, plates 13–36. Tov and I have argued that these five manuscripts—4Q158, 4Q364, 4Q365, 4Q366 and 4Q367—were all copies of a single composition. Recently, Michael Segal has argued that these are separate compositions. See his forthcoming paper, “4QReworked Pentateuch or 4QPentateuch?” in The Dead Sea Scrolls—Fifty Years After Their Discovery—Proceedings of the Jerusalem Congress, July 10–25, 1997, ed. Lawrence H. Schiffman, Tov and VanderKam (Jerusalem: Israel Exploration Society, forthcoming). The resolution of this question is not important for our purposes here.
5.
This is called “innerbiblical exegesis.” See Michael Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon Press, 1985).
6.
This is the position of Ulrich (“The Qumran Scrolls and the Biblical Text,” in Schiffman, Tov and VanderKam, The Dead Sea Scrolls).
7.
The Temple Scroll from Qumran may quote a passage in Leviticus from 4Q365. See Tov and White, “4Q365,” in VanderKam et al., Discoveries in the Judaean Desert XIII, pp. 290–296.
8.
Debates about which books belonged in the canon went on for some time. Esther did not gain universal acceptance until the third century A.D. There are also different canons for different groups: The Hebrew Bible, Septuagint and Old Testament are all different, although with much overlap.
