Written by A standalone piece
Tradition holds that Moses wrote the first five books of the Bible. The text itself makes that impossible — it narrates Moses’ own death and burial (Deuteronomy 34), tells the same stories twice with conflicting details, switches names for God, and contradicts itself on basic points of law and sequence. For more than two centuries, scholars reading those seams have concluded the obvious: the Torah is not one book by one author. It is a composite, stitched from multiple sources by editors over centuries, and given its final shape long after Moses could have lived. The exact number of sources is debated. The composite character is not.
How the seams were found
The detective work began in the Enlightenment. As early as 1711 H. B. Witter, and then the French physician Jean Astruc in 1753, noticed that the text alternates between two names for God — Yahweh and Elohim — and that the switch tends to track other differences in style and content, as if two sources had been braided together.[^1] Through the nineteenth century, scholars including Karl Heinrich Graf and Abraham Kuenen developed the analysis, and Julius Wellhausen gave it its classic form in his 1878 Prolegomena to the History of Israel.[^2]
The classic four-source model
Wellhausen’s version — the Documentary Hypothesis — held that the Pentateuch was woven from four originally independent sources, each with its own vocabulary, theology, and political setting:[^3]
- J (the Yahwist): uses the name Yahweh, vivid and anthropomorphic, dated by Wellhausen to roughly the tenth–ninth century in the southern kingdom of Judah.
- E (the Elohist): uses Elohim until the name Yahweh is revealed to Moses, dated to the ninth–eighth century in the northern kingdom.
- D (the Deuteronomist): essentially the book of Deuteronomy, connected to the reform under King Josiah around 620 BCE — the “book of the law” conveniently discovered in the temple (2 Kings 22).
- P (the Priestly source): the genealogies, ritual law, and ordered creation account, dated to the exilic or post-exilic period, around 500 BCE or later.
These were combined by a succession of redactors — J with E, then JE with D, finally the whole with P — and tradition since Spinoza often named Ezra, the Persian-era leader who re-established the Jerusalem community under Artaxerxes I, as the final editor.[^4] On this scheme, no part of the Torah predates the monarchy, and the book reached its final form only in the Persian period.
What’s solid and what’s contested
Here honesty matters, because the model is often presented as more settled than it is. Two different claims have to be separated.
The claim that the Torah is composite — multiple sources, edited over centuries, finalized late — is robust and mainstream. It rests on the doublets, the contradictions, the divine-name pattern, the anachronisms, and the impossibility of Mosaic authorship, and essentially no critical scholar disputes it.[^5]
The claim that there were exactly four sources, of those dates, combined in that order — Wellhausen’s specific architecture — is not the current consensus. Since the late twentieth century the classic scheme has been heavily critiqued and revised: some scholars favor supplementary models (a core text expanded in layers) or fragmentary models (many small units) over four clean documents; the existence and independence of E is disputed; and the dating of the sources, especially J, is contested.[^6] The terminology and core insights of the Documentary Hypothesis still frame the discussion, but its precise four-source form is one model among several, not settled fact.
So state it the way the evidence allows: the Torah is demonstrably composite and demonstrably late in its final form — that is the strong, unassailable claim. How exactly the sources should be counted and dated is a live scholarly argument, and pretending otherwise hands a critic an easy opening.
The Persian-period shaping
What can be said with confidence about the final stage is the most relevant point for everything else in this series. Whatever the sources, the Pentateuch reached its finished, authoritative form in the Persian period — the same window in which Israelite religion was consolidating its monotheism and acquiring its new eschatology. The Priestly material, with its ordered cosmos and its ritual system, is exilic or post-exilic; the redaction that fused the Torah into a single authoritative document belongs to the era of the return and reconstruction under Persian rule.[^7] Some scholars have gone further and proposed that the codification was encouraged by Persian imperial administration itself — the so-called “imperial authorization” model — though that specific mechanism is debated and should be flagged as a hypothesis rather than a fact.[^8]
The unavoidable consequence is the same one the Moses article reached from the other direction: the foundational law-book of Judaism, traditionally credited to a Bronze Age prophet, is in its actual form a Persian-period compilation — edited, harmonized, and canonized by a community rebuilding its identity centuries after the events it narrates.
The bottom line
The Torah is many hands wearing one name, exactly as Isaiah is. The seams are visible — the doubled stories, the divine-name switch, the law that contradicts itself, the prophet narrating his own death — and they have been visible to anyone willing to read for them since the eighteenth century. The honest, load-bearing claims are that the text is composite and that it was finalized in the Persian period. The one place to keep discipline is the source-counting: “the Torah is composite and late” is bedrock; “there were exactly four sources, J E D P, in that order and those dates” is a particular model now under active revision. Hold the bedrock and you cannot be dislodged. Overstate the architecture and you hand the defense its rebuttal.
Notes
[^1]: On H. B. Witter (c. 1711) and Jean Astruc (1753) noticing the alternation of divine names as evidence of sources, see the standard histories of Pentateuchal criticism; Astruc, Conjectures sur les mémoires originaux (1753).
[^2]: On the development from Graf and Kuenen to Wellhausen, see Julius Wellhausen, Prolegomena zur Geschichte Israels (1878; Eng. Prolegomena to the History of Israel).
[^3]: On the four sources (J, E, D, P) and their characteristic features and Wellhausen’s datings, see the Documentary hypothesis overview and Richard Elliott Friedman, Who Wrote the Bible? (New York: Summit, 1987).
[^4]: On the redactional combination of the sources and the traditional identification of Ezra (the Persian-era leader under Artaxerxes I) as final redactor, see the Documentary hypothesis summaries.
[^5]: On the broad consensus that the Pentateuch is composite, multi-source, and finalized late — distinct from agreement on the specific scheme — see David M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (Oxford University Press, 2011).
[^6]: On the late-twentieth-century critiques and the rise of supplementary and fragmentary models, the dispute over E, and contested datings, see Carr, The Formation of the Hebrew Bible, and the survey in the Documentary hypothesis literature noting the hypothesis has been “critiqued and challenged by other models, especially in the last part of the 20th century.”
[^7]: On the exilic/post-exilic dating of the Priestly material and the Persian-period redaction of the Pentateuch into its final form, see Carr, The Formation of the Hebrew Bible, and standard introductions.
[^8]: On the “Persian imperial authorization” hypothesis (associated with Peter Frei) and the debate around it, see Joseph Blenkinsopp’s and others’ discussions; it is presented here as a contested proposal, not an established mechanism.