You are here

Who Wrote the Torah?

Textual, Historical, Sociological, and Ideological Cornerstones of the Formation of the Pentateuch

By Konrad Schmid Published 2018

Image by Steven G. Johnson, CC BY-SA 3.0
King Jehu of Israel, Black Obelisk of Shalmaneser III, second half of ninth century B.C.E.

Who wrote the Torah? In light of more than two hundred years of scholarship and of the ongoing disputes on that question,[1] the most precise answer to this question still is: We don’t know. The tradition claims it was Moses, but the Torah itself says otherwise. Only small portions within the Torah are traced back to him, but not nearly the whole Torah: Exodus 17:14 (Battle against Amalek); 24:4 (Covenant Code); 34:28 (Ten Commandments); Numbers 33:2 (Wandering Stations); Deuteronomy 31:9 (Deuteronomic Law); and 31:22 (Song of Moses). Despite all disagreement in current scholarship, however, the situation in Pentateuchal research is far from desperate, and there are indeed some basic statements that can be made regarding the formation of the Torah. This is what this contribution is about. It is structured in the following three parts: the textual evidence of the Pentateuch; the socio-historical conditions for the development of the Pentateuch, and “Ideologies” or “Theologies” of the Pentateuch in their historical contexts.

The Textual Evidence of the Pentateuch

What is the textual basis for the Pentateuch?[2] What are the oldest manuscripts we have? At this point, one should mention the so-called Codex Leningradensis or B 19A in the first place.[3] This manuscript of the Hebrew Bible dates to the year 1008 C.E., so it is a medieval text, but it is the oldest complete textual witness to the Pentateuch. This seems to leave us in a very awkward position: We are dealing with an allegedly 2500-year-old text, but its earliest textual attestation is only 1000 years old. Yet the situation is not hopeless.

Firstly, there are ancient translations that significantly predate the Codex B 19 A. The first are the big codices of the translation of the Hebrew Bible into Greek, the earliest of which is the Codex Sinaiticus.[4] While this text is not an original, it is a good witness to the Hebrew text behind it, dating from the fourth century C.E. The Greek text of the Pentateuch shows differences from the Hebrew text, particularly in Exodus 35–40. This issue was noted in 1862 by Julius Popper, who was the first to deal extensively and deliberately with post-Persian expansions in the Pentateuch.[5]

Secondly, there are older, preserved portions of the Pentateuch in Hebrew. Before 1947, the oldest extant fragment of a biblical text was the so-called Papyrus Nash, which probably dates around 100 B.C.E. and contains both the Decalogue and the beginning of the “Shema Israel” from Deuteronomy 6.[6]

Much more important were the textual discoveries from the Dead Sea that began in 1947.[7] Remants of about 900 scrolls were discovered, among them many biblical texts. They date mainly from the second and first centuries B.C.E. Most of the texts are fragmentary, many of them no larger than a few square centimeters. All of the biblical fragments are accessible in Eugene Ulrich’s book The Biblical Qumran Scrolls.[8]

What do these Qumran texts reveal about the Pentateuch in the early, post-biblical period? The most important insight is the remarkable closeness of these fragments, as far as they have been preserved, to the Codex B 19 A. In the case of Gen 1:1–5 in 4QGenb, no differences are present at all.[9]

Nevertheless, the various scrolls seem to display affilitations to the traditionally known, post-70 C.E. textual families of the Pentateuch. Armin Lange gives the following estimate:[10]

Proto-Masoretic: 37.5%
Proto-Samaritan: 5.0%
Proto-Septuagint: 5.0 %
Independent: 52.5 %

In these figures, there is some prevalence of the proto-MT strand, though one observes a significant number of independent readings. At times, the differences are quite relevant, such as the reading of “Elohim” instead of “Yhwh” in Genesis 22:14,[11] or of “Mount Gerizim” instead of “Mount Ebal” in Deuteronomy 27:4 (but the latter fragment might be a forgery).[12] Regarding the large portion of proto-Masoretic texts, Emanuel Tov maintained:

“The differences between these texts [cs. the proto-MT texts] and L [sc. Codex Leningradensis] are negligible, and in fact their nature resembles the internal differences between the medieval manuscripts themselves.”[13]

Thus, the Qumran findings provide an important starting point for Pentateuchal exegesis and corroborate the legitimacy of critically using MT in Pentateuchal research. On the one hand, we can have considerable confidence in the Hebrew text of the Pentateuch, as attested in the medieval manuscript of the Codex B 19A, which is the textual basis for most modern Bible editions. On the other hand, at the time, there was apparently not a fully stable text of the Pentateuch in terms of every single letter or word being fixed as part of a fully canonized Bible, as the differences between the scrolls show.[14]

In terms of the composition of the Pentateuch, another insight that we can deduce from Qumran is that the Pentateuch was basically finished no later than the second century B.C.E. Some of its texts are certainly much older, but probably none of them are later.

One epigraphical piece relating to our concerns should be mentioned: There is a quasi-biblical text from biblical times, the silver amulets from Ketef Hinnom, which offer a text close to Numbers 6:24–26 and date anywhere between the seventh and the second century B.C.E., but this is not really a witness to the Bible.[15]

Socio-Historical Conditions for the Development of the Pentateuch

How should we imagine the cultural-historical background of the Pentateuch’s composition? A very insightful book by Christopher Rollston brings together all of the relevant evidence regarding writing and literacy in ancient Israel.[16] In addition, Matthieu Richelle and Erhard Blum have recently published important contributions that fairly evaluate the evidence of scribal activities in early Israel and Judah.[17]

The first question here is, who could actually read and write? We have different estimates for the ancient world, but they agree that probably not more than 5–10% of the population were literate to a degree that they could read and write texts of some length. Literacy was probably an elite phenomenon, and texts were circulated only among these circles, which were centered around the palace and the temple.[18] In biblical times, producing literature was an enterprise mainly restricted to professional scribes, and reading literature was generally limited to the same circles that produced it.

Recently, Israel Finkelstein and others have claimed that the Lakhish-Ostraca show at least six different scripts, pointing to more widespread literacy even among soldiers in the early sixth century B.C.E.[19] But this kind of evidence remains debatable.

Othmar Keel,[20] Matthieu Richelle[21] and others have argued for a continuous literary tradition in Jerusalem from the Bronze Age city state to the early Iron Age. While this perspective is probably not entirely wrong, it should not be overestimated. Abdi-Hepa’s Jerusalem was something different from David or Solomon’s Jerusalem, and there was obviously a cultural break between Late Bronze and early Iron Age Jerusalem. A case in point would be the new Ophel inscription from Jerusalem, which exhibits a rather rudimentary level of linguistic education.[22]

A second question is, How did people write? Most of the inscriptions we have are on potsherds or stone, but this only what has survived. For obvious reasons, texts on stone or clay last much longer than those on papyrus or leather, so we cannot simply extrapolate from what archeologists have found to what people wrote on in general. (In fact, there is only a single papyrus sheet left from the time of the monarchy, Mur. 17).[23] In addition, we have an impressive number of seals and bullae from Jerusalem during the First Temple period with remnants of papyrus on them that prove that papyrus was a common medium for writing. Some of the bullae bear names such as “Gemaryahu ben Shafan,” who is mentioned in Jeremiah 36:10, or “Yehuchal Ben Shelamayahu” and "Gedaliah Ben Pashchur,” whom we know from Jeremiah 38:1.[24]

In all likelihood, the writing material for texts such as those in the Pentateuch was papyrus or leather: Longer books needed to be written on leather, because papyrus sheets are fragile. The ink was composed of grime and metal. Scholars estimate that it took a professional scribe six months to copy a book the length of Genesis or Isaiah. If one adds the value of the sheep’s skins, it is evident how costly the production of such a scroll would have been.

In biblical times, copies of the books of the Bible were probably very few in number. For the second century B.C.E., 2 Maccabees 2:13–15 provides evidence that the Jewish community in Alexandria, likely among the largest diaspora groups, did not possess a copy of every biblical book. This text quotes a letter from the Jerusalemites to the Jews in Alexandria that invites them to borrow a copy of those biblical books from Jerusalem that they do not possess.

“Nehemiah … founded a library and collected the books about the kings and prophets, and the writings of David…. In the same way Judas [Maccabaeus] also collected all the books that had been lost on account of the war that had come upon us, and they are in our possession. So if you have need of them, send people to get them for you.” (2 Maccabees 2:13–15)

But when was the Pentateuch was composed? It is helpful at the outset to determine a time span in which its texts were written. For the terminus a quo, an important clarification is needed. We can only determine the beginnings of the earliest written versions of a text. In other words, this does not include a text’s oral prehistory. Many texts in the Bible, especially in the Pentateuch, go back to oral traditions that can be much older than their written counterparts. Therefore, the terminus a quo only determines the beginning of the written transmission of a text which, in turn, may have already been known as an oral tale or the like.[25]

Unlike many prophetic texts, Pentateuchal texts do not mention dates of authorship. One must therefore look for internal and external indicators in order to determine the date of their composition.

There is a basic observation relevant for determining the beginnings of the Pentateuch’s literary formation. We can safely determine a historical break in the ninth and eighth centuries B.C.E. in the cultural development of Israel and Judah. This point holds despite Richelle and Blum,[26] who provide sufficient evidence to include the late ninth century as the beginning of this watershed with regard to the development of Israel’s and Judah’s scribal culture. By this point, a certain level of statehood and literacy was being achieved, and these two elements go together. That is, the more developed a state, the more bureaucracy and education are needed—especially in the area of writing.

When one considers the number of inscriptions found in ancient Israel and Judah, the numbers clearly increase in the eighth century, and this increase should probably be interpreted as indicating a cultural development in ancient Israel and Judah. This claim can be corroborated by looking at the texts that have been found and that can be dated to the tenth century B.C.E., such as the Gezer Calendar;[27] the potsherd from Jerusalem;[28] the Baal inscription from Bet Shemesh;[29] the Tel Zayit Abecedary;[30] and the Qeiyafa ostracon.[31] All of them stem from or around the tenth century B.C.E. The modesty of their content and writing style alike are easy to discern.

If we move forward about one century to the ninth century B.C.E., then the evidence is much more telling, even if some of the evidence is in Aramaic and not Hebrew. The first monumental stela from the region is the Mesha Stela, which is written in Moabite and which contains the first documented reference to Yhwh and Israel as we know them.[32] Another monumental text is the Tel Dan stela in Aramaic, best known for mentioning the “Beth David.”[33]

Still another piece of evidence is the eighth-century Aramaic wall inscription from Tell Deir Allah,[34] which mentions the prophet Balaam that appears in Numbers 22–24. Balaam’s story in the inscription is completely different from the narrative about him in the Bible, yet it remains one of the earliest piece of evidence for a literary text in the near vicinity of ancient Israel.

Along with others, Erhard Blum has recently argued convincingly for interpreting the site of Tell Deir Allah as a school, because of a late Hellenistic parallel to the building architecture of Trimithis in Egypt (ca. fourth century C.E.).[35] This interpretation as a school might also be true for Kuntillet Ajrud, where we also have writings on the wall.[36]

The landmark set in the ninth and eighth century B.C.E. by the high amount and new quality of written texts in ancient Israel and Judah corresponds to another relevant feature. At this time, Israel begins to be perceived by its neighbors as a state. That is, not only internal changes in the development of writing, but also external, contemporaneous perceptions hint that Israel and Judah had reached a level of cultural development in the eighth–ninth centuries to enable literary text production.

King Asarhaddon and his subdues, Victory Stele of Esarhaddon, ca. 670 B.C.E.

A good example are the Assyrian inscriptions from the mid-ninth century B.C.E. that mention Jehu, the man of Bit-Humri, which means Jehu of the house of Omri. The Black Obelisk even displays Jehu in a picture (bowing in front of the Assyrian king), being the oldest extant image of an Israelite.[37]

Based on these observations about the development of a scribal culture in ancient Israel, we can assume that the earliest texts in the Pentateuch may have originated as literary pieces from the ninth and eighth centuries B.C.E. But to repeat: This chronological claim pertains only to their literary shape, whereas the oral traditions behind them could be much older, perhaps at times reaching back into the second millennium B.C.E.

When was the Pentateuch finished? On this matter, three areas of evidence should be named. First, there is the translation into Greek, the so-called Septuagint, which can be dated to the mid-second century B.C.E.[38] There are some differences, especially in the second tabernacle account of Exodus 35–40,[39] but the Septuagint basically points to a completed Pentateuch. Secondly, the books of Chronicles and Ezra-Nehemiah, which probably date to the fourth century B.C.E., refer to a textual body called either the Torah of YHWH or the Torah of Moses. It is not clear whether this denotes an already completed Pentateuch, but it at least points in this direction.[40] Thirdly, the Pentateuch has no clear allusion to the Persian empire’s fall in the wake of Alexander the Great’s conquests.[41] The Persian empire lasted from 539–333 B.C.E., a period perceived in ancient Israel as one of political stability—in some texts even marking the end of history. The loss of this political order was accompanied by numerous questions. Especially in Prophetic literature, this event was interpreted as a cosmic judgment. But in the Pentateuch, no text seems to allude to the event directly or indirectly. Therefore, the Pentateuch seems basically to be a pre-Hellenistic text, pre-dating Alexander the Great and the Hellenization of the East.

However, there are a few exceptions to the pre-Hellenistic origins of the Pentateuch. The best candidate for a post-Persian, Hellenistic text in the Pentateuch seems to be the small “apocalypse” in Numbers 24:14–24, which in verse 24 mentions the victory of the ships of the כתים over Ashur and Eber. This text seems to allude to the battles between Alexander and the Persians, as some scholars have suggested.[42] Other post-Persian elements might be the specific numbers in the genealogies of Genesis 5 and 11.[43] These numbers build the overall chronology of the Pentateuch and differ significantly in the various versions. But these exceptions are minor. The substance of the Pentateuch seems to be pre-Hellenistic.

“Ideologies” or “Theologies” of the Pentateuch in Their Historical Contexts

If we can assume with some probability that the Pentateuch was written between the ninth and the fourth centuries B.C.E., how can we reconstruct its literary genesis in greater detail? We should begin by introducing a very general observation. Ancient Israel is part of the ancient Near East. Ancient Israel was a small political entitiy surrounded by greater, and much older, empires in Egpyt and Mesopotamia. It is therefore more than likely that Israel’s literature was deeply influenced by its neighbours and their ideologies and theologies. An extraordinary piece of evidence of cultural transfer is a fragment of the Gilgamesh epic (dating to the fourteenth century B.C.E.) found in Megiddo in northern Israel. The fragment proves that Mesopotamian literature was known and read in the Levant. Also noteworthy is the text of Darius’s late-sixth-century Behistun inscription both in Persia and in Egypt, where it existed as an Aramaic translation.

Of course, there are indigenous traditions in ancient Israel that are not paralleled in other ancient Near Eastern material. But some of the most prominent texts in the Pentateuch creatively adapt the ancient world’s knowledge, and it is important to discern this background in order to understand the biblical texts properly and with their own emphases.

Addressing this topic exhaustively is not possible at the moment. Instead, I’ll pick out two well known examples to demonstrate how prominent biblical texts arose as receptions and adaptions of ancient Near Eastern imperial ideologies. That does not mean that the Bible is not an original text. What it does mean is that the Bible’s originality and creativity are not necessarily to be found in the materials it contains, but in the interpretive adaptations that it applies to these materials.

The first example of how the ancient Near East shaped the Pentateuch has to do with the Neo-Assyrian empire, the preeminent power in the ancient world of the ninth–seventh centuries B.C.E.[44] Its ideology was based on the strict submission of the Assyrian king’s subordinates, as portrayed in this image: Here, the Assyrian king is the master, and all other kings are to serve him.

The Dome of Syracuse, Italy, formerly a Greek Temple (fifth century B.C.E.) and a mosque (ninth century C.E.)

The Assyrians secured their power through treaties with their vassals. These treaties are usually have a three-part structure, containing an introduction, a corpus of stipulations, and a concluding section with blessings and curses.

It is noteworthy that the book of Deuteronomy exhibits this same structure, apparently having been shaped according to the model of an Assyrian vassal treaty. But there is one big difference: The function of Assyrian vassal treaties was to oblige subdued people to the Assyrian king in terms of absolute loyalty. The book of Deuteronomy likewise demands absolute loyalty from the people of Israel, but to God, not to the Assyrian king.

So the book of Deuteronomy seems to take up both the structure and the basic concept of an Assyrian vassal treaty, while at the same time reinterpreting it. With Eckart Otto, Thomas Römer, Nathan Macdonald and others,[45] we therefore can maintain that at least a core of Deuteronomy originated in the late Neo-Assyrian Period in an anti-Assyrian milieu of scribes.

A second example of how the ancient Near East shaped the Pentateuch has to do with the Persian empire. In 539 B.C.E., the Babylonian empire was overthrown by the Persians, after which the Persians ruled the entire ancient world as it was known in that part of the globe for the next two hundred years. Persian rule was perceived by many peoples in the Levant as peaceful, with the era seen as a quiet one, where various peoples could live according to their own culture, language, and religion. In the Hebrew Bible, nearly every foreign nation is addressed with very harsh curses except for the Persians, probably due to their tolerant policy towards those whom they subdued.

In the Pentateuch, we can locate some indications of Persian imperial ideology. A very telling piece is the so-called table of nations in Genesis 10. This text explains the order or the world after the flood, and it structures the seventy people of the globe according to the offspring of Shem, Ham, and Japheth, including three, nearly identical refrains:[46]

בני יפת [...] בארצתם אישׁ ללשׁנו למשׁפחתם בגויהם
Genesis 10:2,5: The sons of Japheth […] in their lands, with their own language, by their families, by their nations.

אלה בני־חם למשׁפחתם ללשׁנתם בארצתם בגויהם
Genesis 10:20: These are the sons of Ham, by their families, by their languages, in their lands, and by their nations.

אלה בני־שׁם למשׁפחתם ללשׁנתם בארצתם לגויהם
Genesis 10:31: These are the sons of Shem, by their families, by their languages, in their lands, and by their nations.

At first glance, these texts may not look very interesting. But they are quite revolutionary insofar as they tell us that the world is ordered in a pluralistic way. After the flood, God intended humanity to live in different nations, with different lands and different languages. Genesis 10 is probably a Persian period text reflecting this basic conviction of Persian imperial ideology. The same ideology is also attested, e.g., in the Behistun inscription, which was disseminated widely throughout the Persian empire.[47]The Persian imperial inscriptions declare that every nation belongs to their specific region and has their specific cultural identities (cf. DNa 30–38; XPh 28–35; DB I 61–71). This structure results from the will of the creator deity, as Klaus Koch pointed out in his “Reichsidee und Reichsorganisation im Perserreich,” where he identifies this structure as “Nationalitätenstaat als Schöpfungsgegebenheit.”[48] Every people should live according to its own tradition and in its own place. This is a radically different political view when compared to the Assyrians and Babylonians, both of whom strove to destroy other national identities, especially by means of deportation. The Persians deported no one, and they allowed people to rebuild their own sanctuaries, such as the temple in Jerusalem that the Babylonians had destroyed.

Once again, though, Genesis 10 is not merely a piece of Persian imperial propaganda. It also includes important interpretive changes. Specifically, it is not the Persian king who determines world order; rather, the God of Israel allots every nation its specific place and language. Of course, the Pentateuch eventually makes clear that Israel has a specific function in the world, but it is important to see that the Bible acknowledges and allows cultural and religious variety in the world.

These examples highlight how the Bible interacts with imperial ideologies from the ancient Near East, a point that is crucial to see if we are to reconstruct its formation.

But how do such different ideologies and theologies go together in the Bible? It is important to see that the Pentateuch in particular and the Bible in general are not uniform pieces of literature. They instead resemble a large cathedral that has grown over centuries. Its content is not the result of one, but rather of many voices. And these different voices establish the overall beauty and richness of the Pentateuch.

Konrad Schmid is Professor of the Hebrew Bible and Ancient Judaism at the University of Zurich. The literary history of the Pentateuch and the reconstruction of the redactional processes that led to its final shape constitute the main focus of his research. Before coming to Zurich, he served as Professor of the Hebrew Bible at the University of Heidelberg (1999–2002). In 2006–07 he was a Member in Residence at the Center for Theological Inquiry in Princeton. In 2012–13, he co-directed a research group on the formation of the Pentateuch at the Israel Institute of Advanced Study, Hebrew University of Jerusalem. In the fall term of 2017, he is a Member of the Institute for Advanced Study in Princeton. His main publications include Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels in den Geschichtsbüchern des Alten Testaments (1999), English translation: Genesis and the Moses Story (2010); and Literaturgeschichte des Alten Testaments: Eine Einführung (2008), translated into English, Portuguese, and Japanese.